Nutrition Reviews

Heileson, Jeffery, L

doi: 10.1093/nutrit/nuz091pmid: 31841151

Abstract The American Heart Association (AHA) recently published a meta-analysis that confirmed their 60-year-old recommendation to limit saturated fat (SFA, saturated fatty acid) and replace it with polyunsaturated fat to reduce the risk of heart disease based on the strength of 4 Core Trials. To assess the evidence for this recommendation, meta-analyses on the effect of SFA consumption on heart disease outcomes were reviewed. Nineteen meta-analyses addressing this topic were identified: 9 observational studies and 10 randomized controlled trials. Meta-analyses of observational studies found no association between SFA intake and heart disease, while meta-analyses of randomized controlled trials were inconsistent but tended to show a lack of an association. The inconsistency seems to have been mediated by the differing clinical trials included. For example, the AHA meta-analysis only included 4 trials (the Core Trials), and those trials contained design and methodological flaws and did not meet all the predefined inclusion criteria. The AHA stance regarding the strength of the evidence for the recommendation to limit SFAs for heart disease prevention may be overstated and in need of reevaluation. heart disease, low-density lipoprotein cholesterol, meta-analysis, polyunsaturated fat, saturated fat, trans fat INTRODUCTION The 2015 Dietary Guidelines for Americans (DGA) recommend consuming <10% of calories from saturated fat (SFA) for optimal health,1 while the American Heart Association (AHA) recommends 5%–6% of total calories as SFA for those at risk of heart disease to reduce low-density lipoprotein cholesterol (LDL-c).2 International food and nutrient guidelines tend to mirror the DGA recommendation.3–5 In contrast to this widespread acceptance, several recent meta-analyses have challenged the evidence and rationale supporting the current dietary recommendations on SFA.6–8 The consensus on SFA came into question after Siri-Tarino et al9 published a meta-analysis of 21 prospective cohort studies of over 347 000 subjects that found no evidence that saturated fat intake is associated with coronary heart disease (CHD) or cardiovascular disease (CVD). Two recent meta-analyses, which included unpublished data, have also challenged the advice to replace SFA with polyunsaturated fat (PUFA).6,10 Despite the shift in the nutritional landscape, the AHA released a Presidential Advisory on dietary fats with a meta-analysis that provided evidence that re-established their nearly 6-decade-old dietary advice to reduce SFA and replace it with PUFA.11 Except for the universally recognized detrimental role of trans fatty acids (TFAs), the dietary fat debate has led to extensive public confusion and questioned the credibility of the DGA.12 In light of this, it is critical to objectively analyze the most recent meta-analytical evidence to determine the role of SFA in heart disease. However, at the outset, it is important to appreciate and understand the genesis and evolution of the current SFA guidelines. Historical timeline of the current saturated fat guidelines In the early 1900s, the relationship between diet and heart disease was first established when Anitschkow and Chalatow13 observed that rabbits fed high-cholesterol diets had elevated serum cholesterol and developed arterial lesions similar to atherosclerosis in humans. In contrast, humans fed high-cholesterol diets failed to elicit a significant increase in serum cholesterol levels.14 When dietary cholesterol appeared an unlikely driver of high serum cholesterol levels and heart disease, in 1950 Gofman et al15 presented data that showed restricting both dietary fat and cholesterol reduced LDL-c in 17 of 20 patients. In 1952, Kinsell et al16 replicated this finding by substituting vegetable oil for animal fat in participants’ diets, dramatically decreasing their serum cholesterol. To explore this relationship further, in 1953 Keys17 compiled dietary and vital statistics from 6 countries and found a linear relationship between dietary fat intake and heart disease mortality. Subsequent epidemiological data indicated a correlation between total intake of dietary fat, serum cholesterol levels, and prevalence of heart disease.18,19 However, in 1970, Keys published the Seven Countries Study, which failed to show an association between total dietary fat and heart disease.20 Rather, this study found that populations with the greatest SFA intake had the highest incidence of heart disease. Thus, the focus of dietary interventions to prevent heart disease shifted from total fat to saturated fat intake. Results of these experimental and observational studies formed the framework for the theory known as the diet-heart hypothesis. This hypothesis postulates that heart disease risk is reduced by limiting saturated fat intake through its ability to lower LDL-c. The hypothesis quickly gained acceptance among clinicians at the time, leading to the AHA’s official recommendation in 1961 to reduce total dietary fat and saturated fat and substitute dietary saturated fat with polyunsaturated fat.21 These recommendations were the foundation for the national dietary guidelines that advocated full-fat dairy products with low- or non-fat versions, fattier cuts of meat with lean meats, butter with margarine, and animal fats and tropical oils with vegetable oils. Recently, these long-standing dietary recommendations have been challenged. Compelling evidence aggregated since 2009 suggests a reevaluation of these recommendations may be warranted.9,22,23 Despite the 2015 Dietary Guidelines for Americans Committee (DGAC) responding to current evidence by deemphasizing cholesterol as a nutrient of concern and removing their recommendation to limit total fat, they did not change their recommendations on SFA.24,25 In response, a panel of nutrition experts from the Academy of Nutrition and Dietetics called for consistent language among expert organizations regarding the DGA stance on SFAs, stating, “In the spirit of the 2015 DGAC’s commendable revision of previous DGAC recommendations to limit dietary cholesterol, the Academy suggests that HHS [the US Department of Health and Human Studies] and USDA [the US Department of Agriculture] support a similar revision deemphasizing saturated fat as a nutrient of concern.”26 More recently, a group of researchers challenged the World Health Organization’s proposed saturated fat recommendation to reduce intake to less than 10% of total calories as this simplistic measure does not take into account the complexity of the food matrix.27 This narrative review aims to provide a brief overview of the recent meta-analyses of dietary SFA and heart disease with special emphasis and review of the methodological approach used by the AHA’s Presidential Advisory on dietary fats and cardiovascular disease (referred to here as the Advisory),11 to include an in-depth review and critique of the 4 clinical trials included. LITERATURE SEARCH STRATEGY PubMed was searched using the terms “dietary saturated fat” AND “heart disease” from 2009 to April 2019. This timeframe was selected since meta-analytical evidence refuting the diet-heart hypothesis began to appear around this time, leading to an increase in meta-analyses of observational studies and randomized controlled trials (RCTs). The search yielded 5287 publications. Once filtered by the article type (meta-analysis and systematic review), 127 articles were available for review. Articles were screened and removed if they were duplicates or systematic reviews without a meta-analysis. Articles were retained if the meta-analysis included studies that compared dietary SFA (high vs low) intake or studies where SFA was replaced with polyunsaturated fat and included data on hard clinical endpoints such as total mortality, coronary heart disease morality, and events; however, omega-3-PUFA–only meta-analyses were not included. Nineteen meta-analyses were identified: 9 observational and 10 randomized controlled trials. Additionally, reference lists from available meta-analyses, 2 Cochrane reviews,28,29 and the Advisory11 were consulted to ensure all relevant studies were captured. META-ANALYSES OF OBSERVATIONAL STUDIES Since 2009, all 9 meta-analyses of observational studies related to dietary fat and heart disease found that SFA intake was not independently associated with heart disease (Table 1).7,9,23,30–35 Two of these studies concluded that replacement of SFA with PUFA was associated with a lower risk of heart disease.30,31 Table 1 Findings from meta-analyses of observational studies of saturated fat and heart disease Reference . Participants (no. of studies) . Author’s conclusion . Jakobsen et al (2009)30 344 696 (11) Replacement of SFA with PUFA decreased CHD events and mortality; however, replacement with MUFA or carbohydrates increased CHD events. Mente et al (2009)34 160 673 (11) Higher intake of SFAs were not associated with CHD events or mortality. Skeaff et al (2009)33 159 433 (7) Noted multiple methodological flaws in dietary fat research. They concluded that “the available evidence from cohort and randomized controlled trials is unsatisfactory and unreliable to make judgement about and substantiate the effects of dietary fat on risk of CHD.” Siri-Tarino et al (2010)9 347 747 (16) No significant evidence that SFAs are associated with increased risk of CHD. Chowdhury et al (2014)32 283 963 (20) No association with dietary or circulating SFAs with CHD. Farvid et al (2014)31 310 602 (13) In prospective observational studies, dietary LA intake is inversely associated with CHD risk in a dose-response manner. These data provide support for current recommendations to replace saturated fat with polyunsaturated fat for primary prevention of CHD. de Souza et al (2015)7 145 922 (6) SFAs are not associated with all-cause mortality or CHD. Harcombe et al (2017)23 89 801 (7) Meta-analysis of prospective cohort studies finds no significant association between CHD deaths and saturated fat consumption. Zhu et al (2019)35 1 302 057 (56) This current meta-analysis of cohort studies suggested that total fat, SFA, MUFA, and PUFA intake were not associated with the risk of cardiovascular disease. Reference . Participants (no. of studies) . Author’s conclusion . Jakobsen et al (2009)30 344 696 (11) Replacement of SFA with PUFA decreased CHD events and mortality; however, replacement with MUFA or carbohydrates increased CHD events. Mente et al (2009)34 160 673 (11) Higher intake of SFAs were not associated with CHD events or mortality. Skeaff et al (2009)33 159 433 (7) Noted multiple methodological flaws in dietary fat research. They concluded that “the available evidence from cohort and randomized controlled trials is unsatisfactory and unreliable to make judgement about and substantiate the effects of dietary fat on risk of CHD.” Siri-Tarino et al (2010)9 347 747 (16) No significant evidence that SFAs are associated with increased risk of CHD. Chowdhury et al (2014)32 283 963 (20) No association with dietary or circulating SFAs with CHD. Farvid et al (2014)31 310 602 (13) In prospective observational studies, dietary LA intake is inversely associated with CHD risk in a dose-response manner. These data provide support for current recommendations to replace saturated fat with polyunsaturated fat for primary prevention of CHD. de Souza et al (2015)7 145 922 (6) SFAs are not associated with all-cause mortality or CHD. Harcombe et al (2017)23 89 801 (7) Meta-analysis of prospective cohort studies finds no significant association between CHD deaths and saturated fat consumption. Zhu et al (2019)35 1 302 057 (56) This current meta-analysis of cohort studies suggested that total fat, SFA, MUFA, and PUFA intake were not associated with the risk of cardiovascular disease. Abbreviations: CHD, coronary heart disease; LA, linoleic acid; MUFA, monounsaturated fatty acids; PUFA, polyunsaturated fatty acids; SFA, saturated fatty acids. Open in new tab Table 1 Findings from meta-analyses of observational studies of saturated fat and heart disease Reference . Participants (no. of studies) . Author’s conclusion . Jakobsen et al (2009)30 344 696 (11) Replacement of SFA with PUFA decreased CHD events and mortality; however, replacement with MUFA or carbohydrates increased CHD events. Mente et al (2009)34 160 673 (11) Higher intake of SFAs were not associated with CHD events or mortality. Skeaff et al (2009)33 159 433 (7) Noted multiple methodological flaws in dietary fat research. They concluded that “the available evidence from cohort and randomized controlled trials is unsatisfactory and unreliable to make judgement about and substantiate the effects of dietary fat on risk of CHD.” Siri-Tarino et al (2010)9 347 747 (16) No significant evidence that SFAs are associated with increased risk of CHD. Chowdhury et al (2014)32 283 963 (20) No association with dietary or circulating SFAs with CHD. Farvid et al (2014)31 310 602 (13) In prospective observational studies, dietary LA intake is inversely associated with CHD risk in a dose-response manner. These data provide support for current recommendations to replace saturated fat with polyunsaturated fat for primary prevention of CHD. de Souza et al (2015)7 145 922 (6) SFAs are not associated with all-cause mortality or CHD. Harcombe et al (2017)23 89 801 (7) Meta-analysis of prospective cohort studies finds no significant association between CHD deaths and saturated fat consumption. Zhu et al (2019)35 1 302 057 (56) This current meta-analysis of cohort studies suggested that total fat, SFA, MUFA, and PUFA intake were not associated with the risk of cardiovascular disease. Reference . Participants (no. of studies) . Author’s conclusion . Jakobsen et al (2009)30 344 696 (11) Replacement of SFA with PUFA decreased CHD events and mortality; however, replacement with MUFA or carbohydrates increased CHD events. Mente et al (2009)34 160 673 (11) Higher intake of SFAs were not associated with CHD events or mortality. Skeaff et al (2009)33 159 433 (7) Noted multiple methodological flaws in dietary fat research. They concluded that “the available evidence from cohort and randomized controlled trials is unsatisfactory and unreliable to make judgement about and substantiate the effects of dietary fat on risk of CHD.” Siri-Tarino et al (2010)9 347 747 (16) No significant evidence that SFAs are associated with increased risk of CHD. Chowdhury et al (2014)32 283 963 (20) No association with dietary or circulating SFAs with CHD. Farvid et al (2014)31 310 602 (13) In prospective observational studies, dietary LA intake is inversely associated with CHD risk in a dose-response manner. These data provide support for current recommendations to replace saturated fat with polyunsaturated fat for primary prevention of CHD. de Souza et al (2015)7 145 922 (6) SFAs are not associated with all-cause mortality or CHD. Harcombe et al (2017)23 89 801 (7) Meta-analysis of prospective cohort studies finds no significant association between CHD deaths and saturated fat consumption. Zhu et al (2019)35 1 302 057 (56) This current meta-analysis of cohort studies suggested that total fat, SFA, MUFA, and PUFA intake were not associated with the risk of cardiovascular disease. Abbreviations: CHD, coronary heart disease; LA, linoleic acid; MUFA, monounsaturated fatty acids; PUFA, polyunsaturated fatty acids; SFA, saturated fatty acids. Open in new tab Jakobsen et al30 conducted a pooled analysis of 11 cohort studies of 344 696 people that examined the effect of replacing saturated fat with monounsaturated fat (MUFA), PUFA, or carbohydrate on heart disease. Jakobsen et al30 concluded that for every 5% replacement of SFA with PUFA there was an associated 13% reduction in CHD events (hazard ratio [HR] 0.87; 95% confidence interval [CI], 0.77–0.97). However, there was an increased risk with replacement of MUFA (HR 1.19; 95%CI, 1.00–1.41) or carbohydrates (HR 1.07; 95%CI, 1.01–1.14).30 Although the study authors noted the dietary exposure as all PUFAs (omega-3 and omega-6), subsequent publications erroneously attributed the decreased risk solely to linoleic acid (LA).36–38 Despite the conflation of terms, the Farvid et al31 analysis of observational studies found that higher LA intake was independently associated with a decrease in CHD events (pooled relative risk [RR], 0.85; 95%CI, 0.78–0.92) and deaths (pooled RR, 0.79; 95%CI, 0.71–0.89), and, albeit weaker, replacement of SFA with LA also reduced CHD events (pooled RR, 0.91; 95%CI, 0.87–0.96). Conversely, Chowdhury et al32 conducted a comprehensive analysis that included cohort and intervention studies, which analyzed omega-6 specifically, and found no significant association with decreased coronary events (RR, 0.98; 95%CI, 0.90–1.06). In agreement, the Mente et al34 meta-analysis of prospective cohort studies found no association between PUFA intake and heart disease (RR, 1.03; 95%CI, 0.81–1.23). Interestingly, Skeaff and Miller33 found that total intake of PUFA (P = 0.009) and LA (P = 0.032) were associated with a 25% increased risk in CHD mortality (RR, 1.25; 95%CI, 1.06–1.47 and RR, 1.25; 95%CI, 1.02–1.52, respectively), but not CHD events (RR, 0.97; 95%CI, 0.74–1.27 – for PUFA). A recent dose-response meta-analysis determined that total fat, MUFA, SFA, and PUFA were not associated with CVD risk.35 Interestingly, in subgroup analysis, SFA intake was found to be significantly associated with reduced CVD risk in an Asian population (HR, 0.84; 95%CI, 0.73–0.97). Also during subgroup analysis, PUFA was inversely associated with CVD risk after 10 years; however, an incremental increase in PUFA (5 g/d) was associated with a 4% increased risk in CVD (HR, 1.04; 95%CI, 1.01–1.07; Plinearity = 0.030). The authors speculated that this could be related to concomitant intake of mercury found in fatty fish. Although plausible, this finding would need to be confirmed, since the extant literature suggests an inverse association with fish consumption and myocardial infarction or cardiovascular mortality, especially in Asian populations.39,40 It should be noted that the replacement of SFA for PUFA, or LA specifically, may not have been observed in the individual cohorts, but statistically determined by nutrient substitution analysis. While there are some benefits to this approach when nutrient replacement occurs isocalorically,41 the substitution analysis does not take into account the accompanying nutrient profile of the foods replaced. For example, a recent study determined that nutrient substitution – dairy fat for PUFA and MUFA – led to a decrease in the nutrient-rich food index; however, a food-level substitution led to an increase in the nutrient-rich food and the healthy eating index (for those with the highest dairy consumption) owing to a reduction in micronutrients such as calcium, vitamin A, vitamin D, and B12.42 Statistical modeling, at the nutrient or food level, adds a new layer of complexity to nutritional epidemiology, but it does not alleviate observational studies of their limited ability to make casual inferences. To better understand whether a relationship exists between the replacement of SFA with PUFA and heart disease, the next section will briefly review the meta-analyses of RCTs. META-ANALYSES OF RANDOMIZED CONTROLLED TRIALS According to the hierarchy of evidence, RCTs offer a way to prove causation from associations determined by observational studies. The observational evidence seems to agree that no independent association exists between saturated fat and heart disease7,9,23,30–34,43; however, 2 of the meta-analyses found an association with replacement of SFA with PUFAs.30,31 Since 2009, 10 meta-analyses of RCTs have been conducted (Table 2).6,8,10,11,22,28,29,44–46 None reported a significant increase in the hard points such as heart disease mortality or total mortality with SFA intake, although 3 found a significant decrease between groups for soft endpoints such as CHD or combined CVD disease events.11,29,44 Table 2 Findings from meta-analyses of randomized controlled trials exploring saturated fat and heart disease Reference . Participants (no. of studies) . Intervention . Author’s conclusion . Mozaffarian et al (2010)44 13 614 (8) Replacement of SFA with PUFA These findings provide evidence that consuming PUFA in place of SFA reduces CHD events in RCTs. Ramsden et al (2010)45 11 275 (7) Replacement of SFA with PUFA Advice to increase n-6 PUFA intake is unlikely to provide the intended benefits and may increase the risks of CHD and death. Hooper et al (2012)28 65 508 (24) Reduced or modified fat This review suggested that reducing saturated fat by reducing and/or modifying dietary fat reduced the risk of CV events. Ramsden et al (2013)10 13 308 (8) Replacement of SFA with mostly LA Selective substitution of n-6 LA for SFA is unlikely to be beneficial, particularly in patients with established CHD. Schwingshackl and Hoffmann (2014)46 7150 (12) SFA vs PUFA (secondary prevention) In secondary CHD prevention, higher intakes of PUFA in replacement of SFA were not associated with risk reduction. Hooper et al (2015)29 59 000 (15) Reduced SFA The findings are suggestive of a small but potentially important reduction in CV risk on reduction of SFA intake. Ramsden et al (2016)6 10 808 (5) Replacement of SFA with LA Replacement of SFA in the diet with LA lowers serum cholesterol but does not support the hypothesis that this lowers risk of death from CHD or all causes. Harcombe et al (2016)22 62 421 (10) Reduced or modified fat Currently available RCT evidence does not support the current dietary fat guidelines. Hamley (2017)8 6045 (5) Replacement of SFA with PUFA (mostly LA) Available evidence from adequately controlled RCT suggest replacing SFA with mostly n-6 PUFA is unlikely to reduce CHD events, CHD mortality, or total mortality. Sacks et al (2017)11 2873 (4) Replacement of SFA with PUFA Lowering SFA and replacing it with vegetable oil rich in PUFA, primarily soybean oil, lowered CHD by 29%. Reference . Participants (no. of studies) . Intervention . Author’s conclusion . Mozaffarian et al (2010)44 13 614 (8) Replacement of SFA with PUFA These findings provide evidence that consuming PUFA in place of SFA reduces CHD events in RCTs. Ramsden et al (2010)45 11 275 (7) Replacement of SFA with PUFA Advice to increase n-6 PUFA intake is unlikely to provide the intended benefits and may increase the risks of CHD and death. Hooper et al (2012)28 65 508 (24) Reduced or modified fat This review suggested that reducing saturated fat by reducing and/or modifying dietary fat reduced the risk of CV events. Ramsden et al (2013)10 13 308 (8) Replacement of SFA with mostly LA Selective substitution of n-6 LA for SFA is unlikely to be beneficial, particularly in patients with established CHD. Schwingshackl and Hoffmann (2014)46 7150 (12) SFA vs PUFA (secondary prevention) In secondary CHD prevention, higher intakes of PUFA in replacement of SFA were not associated with risk reduction. Hooper et al (2015)29 59 000 (15) Reduced SFA The findings are suggestive of a small but potentially important reduction in CV risk on reduction of SFA intake. Ramsden et al (2016)6 10 808 (5) Replacement of SFA with LA Replacement of SFA in the diet with LA lowers serum cholesterol but does not support the hypothesis that this lowers risk of death from CHD or all causes. Harcombe et al (2016)22 62 421 (10) Reduced or modified fat Currently available RCT evidence does not support the current dietary fat guidelines. Hamley (2017)8 6045 (5) Replacement of SFA with PUFA (mostly LA) Available evidence from adequately controlled RCT suggest replacing SFA with mostly n-6 PUFA is unlikely to reduce CHD events, CHD mortality, or total mortality. Sacks et al (2017)11 2873 (4) Replacement of SFA with PUFA Lowering SFA and replacing it with vegetable oil rich in PUFA, primarily soybean oil, lowered CHD by 29%. Abbreviations: CHD, coronary heart disease; CV, cardiovascular; LA, linoleic acid; PUFA, polyunsaturated fatty acids; RCT, randomized controlled trial; SFA, saturated fatty acids. Open in new tab Table 2 Findings from meta-analyses of randomized controlled trials exploring saturated fat and heart disease Reference . Participants (no. of studies) . Intervention . Author’s conclusion . Mozaffarian et al (2010)44 13 614 (8) Replacement of SFA with PUFA These findings provide evidence that consuming PUFA in place of SFA reduces CHD events in RCTs. Ramsden et al (2010)45 11 275 (7) Replacement of SFA with PUFA Advice to increase n-6 PUFA intake is unlikely to provide the intended benefits and may increase the risks of CHD and death. Hooper et al (2012)28 65 508 (24) Reduced or modified fat This review suggested that reducing saturated fat by reducing and/or modifying dietary fat reduced the risk of CV events. Ramsden et al (2013)10 13 308 (8) Replacement of SFA with mostly LA Selective substitution of n-6 LA for SFA is unlikely to be beneficial, particularly in patients with established CHD. Schwingshackl and Hoffmann (2014)46 7150 (12) SFA vs PUFA (secondary prevention) In secondary CHD prevention, higher intakes of PUFA in replacement of SFA were not associated with risk reduction. Hooper et al (2015)29 59 000 (15) Reduced SFA The findings are suggestive of a small but potentially important reduction in CV risk on reduction of SFA intake. Ramsden et al (2016)6 10 808 (5) Replacement of SFA with LA Replacement of SFA in the diet with LA lowers serum cholesterol but does not support the hypothesis that this lowers risk of death from CHD or all causes. Harcombe et al (2016)22 62 421 (10) Reduced or modified fat Currently available RCT evidence does not support the current dietary fat guidelines. Hamley (2017)8 6045 (5) Replacement of SFA with PUFA (mostly LA) Available evidence from adequately controlled RCT suggest replacing SFA with mostly n-6 PUFA is unlikely to reduce CHD events, CHD mortality, or total mortality. Sacks et al (2017)11 2873 (4) Replacement of SFA with PUFA Lowering SFA and replacing it with vegetable oil rich in PUFA, primarily soybean oil, lowered CHD by 29%. Reference . Participants (no. of studies) . Intervention . Author’s conclusion . Mozaffarian et al (2010)44 13 614 (8) Replacement of SFA with PUFA These findings provide evidence that consuming PUFA in place of SFA reduces CHD events in RCTs. Ramsden et al (2010)45 11 275 (7) Replacement of SFA with PUFA Advice to increase n-6 PUFA intake is unlikely to provide the intended benefits and may increase the risks of CHD and death. Hooper et al (2012)28 65 508 (24) Reduced or modified fat This review suggested that reducing saturated fat by reducing and/or modifying dietary fat reduced the risk of CV events. Ramsden et al (2013)10 13 308 (8) Replacement of SFA with mostly LA Selective substitution of n-6 LA for SFA is unlikely to be beneficial, particularly in patients with established CHD. Schwingshackl and Hoffmann (2014)46 7150 (12) SFA vs PUFA (secondary prevention) In secondary CHD prevention, higher intakes of PUFA in replacement of SFA were not associated with risk reduction. Hooper et al (2015)29 59 000 (15) Reduced SFA The findings are suggestive of a small but potentially important reduction in CV risk on reduction of SFA intake. Ramsden et al (2016)6 10 808 (5) Replacement of SFA with LA Replacement of SFA in the diet with LA lowers serum cholesterol but does not support the hypothesis that this lowers risk of death from CHD or all causes. Harcombe et al (2016)22 62 421 (10) Reduced or modified fat Currently available RCT evidence does not support the current dietary fat guidelines. Hamley (2017)8 6045 (5) Replacement of SFA with PUFA (mostly LA) Available evidence from adequately controlled RCT suggest replacing SFA with mostly n-6 PUFA is unlikely to reduce CHD events, CHD mortality, or total mortality. Sacks et al (2017)11 2873 (4) Replacement of SFA with PUFA Lowering SFA and replacing it with vegetable oil rich in PUFA, primarily soybean oil, lowered CHD by 29%. Abbreviations: CHD, coronary heart disease; CV, cardiovascular; LA, linoleic acid; PUFA, polyunsaturated fatty acids; RCT, randomized controlled trial; SFA, saturated fatty acids. Open in new tab Mozaffarian et al44 reviewed 8 RCTs from 7 cohorts to determine whether substitution of SFA with PUFA consumption would have an effect on CHD outcomes. The pooled risk reduction for CHD events was 19% (RR = 0.81; 95%CI, 0.70–0.95; P = 0.008). Unfortunately, their analysis included a nonrandomized controlled trial, the Finnish Mental Hospital Study (FMHS),47,48 and a multifactorial study, the Oslo Diet Heart Study (ODHS),49 and excluded the only study that showed an increase in mortality with the degree of oil saturation – the Rose Corn Oil trial.50 Mozaffarian et al44 referred to the FMHS as a cluster-randomized crossover trial; however, a Cochrane systematic review available at the time excluded the FMHS since it was not a randomized trial.51 After 2010, all meta-analyses of randomized controlled trials excluded the FMHS from their analysis (Table 3),8,10,28,29,45,52 except the most recent Advisory.11 Despite the inclusion of FMHS and ODHS, the authors noted that, “given these limitations of each individual trial, the quantitative pooled risk estimate should be interpreted with some caution.”44 Table 3 The Finnish Mental Hospital Study (FMHS) design issues noted by meta-analysis authors Reference . Author(s) comments . Ramsden et al (2010)45 “… patients were assigned by hospital and not randomized as individual patients …” Hooper et al (2012)28 “We did not include any cluster randomized trials in this review, and cross-over studies (such as the Finnish Mental Hospital Study, 1972) were excluded as this design would be inappropriate for assessing effects on cardiovascular events or mortality.” “… inappropriate in a progressive condition such as cardiovascular disease.” Ramsden et al (2013)10 “… was excluded because patients were assigned by hospital and not randomized as individual patients …” Hooper et al (2015)29 “Not randomized (cluster-randomized, but <6 clusters).” Harcombe et al (2015)52 “… excluded, as they were not randomized. The Finnish study was also a cross-over trial. This is not appropriate for the examination of a long-term mortality, as deaths in the second phase may be due to conditions imposed during the initial phase.” Hamley (2017)8 “Participants were allocated by hospital and were not individually randomized in FMHS, and while it has been suggested to be a cluster randomized trial, there would only have been 2 clusters and there is actually no mention of random allocation of the hospitals in the publications from the trial.” Reference . Author(s) comments . Ramsden et al (2010)45 “… patients were assigned by hospital and not randomized as individual patients …” Hooper et al (2012)28 “We did not include any cluster randomized trials in this review, and cross-over studies (such as the Finnish Mental Hospital Study, 1972) were excluded as this design would be inappropriate for assessing effects on cardiovascular events or mortality.” “… inappropriate in a progressive condition such as cardiovascular disease.” Ramsden et al (2013)10 “… was excluded because patients were assigned by hospital and not randomized as individual patients …” Hooper et al (2015)29 “Not randomized (cluster-randomized, but <6 clusters).” Harcombe et al (2015)52 “… excluded, as they were not randomized. The Finnish study was also a cross-over trial. This is not appropriate for the examination of a long-term mortality, as deaths in the second phase may be due to conditions imposed during the initial phase.” Hamley (2017)8 “Participants were allocated by hospital and were not individually randomized in FMHS, and while it has been suggested to be a cluster randomized trial, there would only have been 2 clusters and there is actually no mention of random allocation of the hospitals in the publications from the trial.” Open in new tab Table 3 The Finnish Mental Hospital Study (FMHS) design issues noted by meta-analysis authors Reference . Author(s) comments . Ramsden et al (2010)45 “… patients were assigned by hospital and not randomized as individual patients …” Hooper et al (2012)28 “We did not include any cluster randomized trials in this review, and cross-over studies (such as the Finnish Mental Hospital Study, 1972) were excluded as this design would be inappropriate for assessing effects on cardiovascular events or mortality.” “… inappropriate in a progressive condition such as cardiovascular disease.” Ramsden et al (2013)10 “… was excluded because patients were assigned by hospital and not randomized as individual patients …” Hooper et al (2015)29 “Not randomized (cluster-randomized, but <6 clusters).” Harcombe et al (2015)52 “… excluded, as they were not randomized. The Finnish study was also a cross-over trial. This is not appropriate for the examination of a long-term mortality, as deaths in the second phase may be due to conditions imposed during the initial phase.” Hamley (2017)8 “Participants were allocated by hospital and were not individually randomized in FMHS, and while it has been suggested to be a cluster randomized trial, there would only have been 2 clusters and there is actually no mention of random allocation of the hospitals in the publications from the trial.” Reference . Author(s) comments . Ramsden et al (2010)45 “… patients were assigned by hospital and not randomized as individual patients …” Hooper et al (2012)28 “We did not include any cluster randomized trials in this review, and cross-over studies (such as the Finnish Mental Hospital Study, 1972) were excluded as this design would be inappropriate for assessing effects on cardiovascular events or mortality.” “… inappropriate in a progressive condition such as cardiovascular disease.” Ramsden et al (2013)10 “… was excluded because patients were assigned by hospital and not randomized as individual patients …” Hooper et al (2015)29 “Not randomized (cluster-randomized, but <6 clusters).” Harcombe et al (2015)52 “… excluded, as they were not randomized. The Finnish study was also a cross-over trial. This is not appropriate for the examination of a long-term mortality, as deaths in the second phase may be due to conditions imposed during the initial phase.” Hamley (2017)8 “Participants were allocated by hospital and were not individually randomized in FMHS, and while it has been suggested to be a cluster randomized trial, there would only have been 2 clusters and there is actually no mention of random allocation of the hospitals in the publications from the trial.” Open in new tab Hooper et al29 reviewed RCTs examining the effects of saturated fat intake over 24 months. Fifteen trials involving 59 000 participants met the inclusion criteria. A reduction in SFA intake had no effect on all-cause, cardiovascular, or CHD mortality; myocardial infarctions; nonfatal myocardial infarctions; stroke; or CHD events; but there was a significant decline in combined CVD events (RR, 0.83; 95%CI, 0.72–0.96). When analyzed by gender, this difference was observed in men (RR, 0.80; 95%CI, 0.69–0.93), but not in women (RR, 1.00; 95%CI, 0.88–1.14). The authors found no benefit to replacing SFA with carbohydrate or protein and found that the effects of replacing it with MUFA were unclear. Replacing SFA with PUFA had no impact on all-cause mortality, cardiovascular mortality, myocardial infarctions, nonfatal myocardial infarctions, stroke, or CHD mortality. There was a reduction in combined CVD events (RR, 0.73; 95%CI, 0.58–0.92). Interestingly, this substitution analysis differs from their earlier meta-analysis.28 In 2012, the Cochrane Collaboration specifically excluded studies with a dietary difference, clear or unclear, in any variable other than fat intake, which excluded the ODHS and the St Thomas Atherosclerosis Regression Study.28 Hooper et al28 noted that both trials were at high risk of bias since the dietary differences—fiber intake from fruits and vegetables and omega-3 PUFAs from fish intake—may have impacted the primary outcomes; however, the most recent review did not exclude the trials despite their selection criteria stating that multifactorial trials would not be included.29 Owing to the substantial between-study heterogeneity observed in the statistically significant findings for combined CVD events in the Hooper et al29 meta-analysis (I2 = 65%), Thornley et al53 reanalyzed the review using an inverse-variance heterogeneity method. This approach was selected in lieu of the random effects model as it maintained the weighted contribution of each trial, potentially reducing publication bias. The updated analysis yielded the same pooled relative risk for combined CVD events (RR 0.93; 95%CI, 0.74–1.16) as the fixed effects model, and was not statistically significant. Only 3 of the 10 meta-analyses found any statistical difference in any CHD or CVD outcome. The divergent conclusions reached may be mediated by the studies included and excluded (Table 4).6,8,10,11,22,28,29,44–46 As discussed previously by Hamley,8 the most common diet-heart trials are the Rose Corn Oil trial, the Oslo Diet-Heart Study (ODHS), the Medical Research Council study, the Los Angeles Veterans Administration trial, the Finnish Mental Hospital Study (FMHS), the Sydney Diet Heart Study (SDHS), the Minnesota Coronary Study (MCS), the Diet and Reinfarction Trial, and the St Thomas Atherosclerosis Regression Study. Only 4 of the trials had an inclusion rate < 80%, most notably the FMHS, which had the lowest inclusion rate at 20% but was included in 2 of the 3 meta-analyses with negative findings on SFA and heart disease and is one of the 4 Core Trials in the Advisory.11 Table 4 The most common diet-heart clinical trials included in meta-analyses of randomized controlled trials exploring saturated fat and heart disease Reference . RCO . ODHS . MRC . LA Vets . FMHS . SDHS . MCS . DART . STARS . Mozaffarian et al (2010)44 X X X X X X X Ramsden et al (2010)45 X X X X X X X Hooper et al (2012)28,a X X X X X X Ramsden et al (2013)10 X X X X X X X Schwingshackl and Hoffmann (2014)46 X X X X X X Hooper et al (2015)29 X X X X X X X X Ramsden et al (2016)6 X X X X X X X X Harcombe et al (2016)22 X X X X X Hamley (2017)8 X X X X X Sacks et al (2017)11 X X X X Inclusion rate (%) 80 80 100 80 20 80 70 60 40 Reference . RCO . ODHS . MRC . LA Vets . FMHS . SDHS . MCS . DART . STARS . Mozaffarian et al (2010)44 X X X X X X X Ramsden et al (2010)45 X X X X X X X Hooper et al (2012)28,a X X X X X X Ramsden et al (2013)10 X X X X X X X Schwingshackl and Hoffmann (2014)46 X X X X X X Hooper et al (2015)29 X X X X X X X X Ramsden et al (2016)6 X X X X X X X X Harcombe et al (2016)22 X X X X X Hamley (2017)8 X X X X X Sacks et al (2017)11 X X X X Inclusion rate (%) 80 80 100 80 20 80 70 60 40 Abbreviations: DART, Diet and Reinfarction Trial; FMHS, Finnish Mental Hospital Study; LA Vets, Los Angeles Veterans Administration trial; MCS, Minnesota Coronary Study; MRC, Medical Research Council; ODHS, Oslo Diet-Heart Study; RCO, Rose Corn Oil trial; SDHS, Sydney Diet-Heart Study; STARS, St Thomas Atherosclerosis Regression Study; X, study included. a Excluded ODHS and STARS during sensitivity analysis. Open in new tab Table 4 The most common diet-heart clinical trials included in meta-analyses of randomized controlled trials exploring saturated fat and heart disease Reference . RCO . ODHS . MRC . LA Vets . FMHS . SDHS . MCS . DART . STARS . Mozaffarian et al (2010)44 X X X X X X X Ramsden et al (2010)45 X X X X X X X Hooper et al (2012)28,a X X X X X X Ramsden et al (2013)10 X X X X X X X Schwingshackl and Hoffmann (2014)46 X X X X X X Hooper et al (2015)29 X X X X X X X X Ramsden et al (2016)6 X X X X X X X X Harcombe et al (2016)22 X X X X X Hamley (2017)8 X X X X X Sacks et al (2017)11 X X X X Inclusion rate (%) 80 80 100 80 20 80 70 60 40 Reference . RCO . ODHS . MRC . LA Vets . FMHS . SDHS . MCS . DART . STARS . Mozaffarian et al (2010)44 X X X X X X X Ramsden et al (2010)45 X X X X X X X Hooper et al (2012)28,a X X X X X X Ramsden et al (2013)10 X X X X X X X Schwingshackl and Hoffmann (2014)46 X X X X X X Hooper et al (2015)29 X X X X X X X X Ramsden et al (2016)6 X X X X X X X X Harcombe et al (2016)22 X X X X X Hamley (2017)8 X X X X X Sacks et al (2017)11 X X X X Inclusion rate (%) 80 80 100 80 20 80 70 60 40 Abbreviations: DART, Diet and Reinfarction Trial; FMHS, Finnish Mental Hospital Study; LA Vets, Los Angeles Veterans Administration trial; MCS, Minnesota Coronary Study; MRC, Medical Research Council; ODHS, Oslo Diet-Heart Study; RCO, Rose Corn Oil trial; SDHS, Sydney Diet-Heart Study; STARS, St Thomas Atherosclerosis Regression Study; X, study included. a Excluded ODHS and STARS during sensitivity analysis. Open in new tab THE AMERICAN HEART ASSOCIATION PRESIDENTIAL ADVISORY ON DIETARY FATS AND CARDIOVASCULAR DISEASE Despite a weak observational foundation, inconsistent RCTs, and 2 recent meta-analyses calling for reevaluation of the current SFA recommendations,6,8 in 2017 the AHA published a position statement that reestablished their 1961 stance on SFA and heart disease.11 Based on 6 inclusion criteria, the AHA identified “4 trials that make up the core evidence on the basis of quality of study design, execution, and adherence” – also known as the 4 Core Trials (4CTs). The trials, listed in order of overall contribution (% weight), are the Finnish Mental Hospital Study (31.66%), the Oslo Diet Heart Study (27.95%), the Medical Research Council study (24.46%), and the Los Angeles Veterans Administration trial (15.93%).11 Despite their inclusion into the Advisory meta-analysis, it appears that of the 4CTs, 3 did not meet the AHA’s prescribed inclusion criteria for dietary control of intake for the intervention and control groups, while all 4 trials failed to the meet the criteria for TFA intake (Table 5).47,49,54,55 The following section provides a discussion of the characteristics, designs, and limitations (primarily the lack of dietary control of the intervention and control groups and dominant confounding variables) of each of the 4CTs, which undermines rather than supports their contention regarding the strength of their evidence for SFA recommendations. Table 5 Inadequate control of confounding variables in the 4 Core Trials (4CTs) 4CTs (year) . TFA intake . Lack of control in diets and other variables . Turpeinen et al (1979)47 SFA group: consumed 9× more margarine SFA group: more cardiotoxic medication, sugar intake, and smoking Overall: transient nature of the patient population led to less than 50% adherence to the trial Leren (1970)49 PUFA group: “highly restricted” shortening, margarine, and hydrogenated oils PUFA group: provided with counseling at hospital and in-home; consumed more fruits, vegetables, legumes, vitamin D, and omega-3 MRC (1968)54 PUFA group “forbidden” to use margarines and cakes Adequately controlled trial Dayton and Pearce (1969)55 PUFA group limited cakes, pastries, and biscuits SFA group: more heavy smokers, fewer nonsmokers, and consumed a vitamin E–deficient diet PUFA group: more than double the attrition; adherence was less than 50% 4CTs (year) . TFA intake . Lack of control in diets and other variables . Turpeinen et al (1979)47 SFA group: consumed 9× more margarine SFA group: more cardiotoxic medication, sugar intake, and smoking Overall: transient nature of the patient population led to less than 50% adherence to the trial Leren (1970)49 PUFA group: “highly restricted” shortening, margarine, and hydrogenated oils PUFA group: provided with counseling at hospital and in-home; consumed more fruits, vegetables, legumes, vitamin D, and omega-3 MRC (1968)54 PUFA group “forbidden” to use margarines and cakes Adequately controlled trial Dayton and Pearce (1969)55 PUFA group limited cakes, pastries, and biscuits SFA group: more heavy smokers, fewer nonsmokers, and consumed a vitamin E–deficient diet PUFA group: more than double the attrition; adherence was less than 50% Abbreviations: MRC, Medical Research Council; PUFA, polyunsaturated fatty acids; SFA, saturated fatty acids; TFA, trans fatty acids. Open in new tab Table 5 Inadequate control of confounding variables in the 4 Core Trials (4CTs) 4CTs (year) . TFA intake . Lack of control in diets and other variables . Turpeinen et al (1979)47 SFA group: consumed 9× more margarine SFA group: more cardiotoxic medication, sugar intake, and smoking Overall: transient nature of the patient population led to less than 50% adherence to the trial Leren (1970)49 PUFA group: “highly restricted” shortening, margarine, and hydrogenated oils PUFA group: provided with counseling at hospital and in-home; consumed more fruits, vegetables, legumes, vitamin D, and omega-3 MRC (1968)54 PUFA group “forbidden” to use margarines and cakes Adequately controlled trial Dayton and Pearce (1969)55 PUFA group limited cakes, pastries, and biscuits SFA group: more heavy smokers, fewer nonsmokers, and consumed a vitamin E–deficient diet PUFA group: more than double the attrition; adherence was less than 50% 4CTs (year) . TFA intake . Lack of control in diets and other variables . Turpeinen et al (1979)47 SFA group: consumed 9× more margarine SFA group: more cardiotoxic medication, sugar intake, and smoking Overall: transient nature of the patient population led to less than 50% adherence to the trial Leren (1970)49 PUFA group: “highly restricted” shortening, margarine, and hydrogenated oils PUFA group: provided with counseling at hospital and in-home; consumed more fruits, vegetables, legumes, vitamin D, and omega-3 MRC (1968)54 PUFA group “forbidden” to use margarines and cakes Adequately controlled trial Dayton and Pearce (1969)55 PUFA group limited cakes, pastries, and biscuits SFA group: more heavy smokers, fewer nonsmokers, and consumed a vitamin E–deficient diet PUFA group: more than double the attrition; adherence was less than 50% Abbreviations: MRC, Medical Research Council; PUFA, polyunsaturated fatty acids; SFA, saturated fatty acids; TFA, trans fatty acids. Open in new tab Finnish Mental Hospital Study The FMHS has been recognized as “one of the earliest and most convincing examples of the better efficacy of unsaturated than of saturated fat in reducing … heart disease.”56 It was initiated in 1958 as a 12-year intervention trial that compared a diet high in PUFA with a diet high in SFA in 1222 patients from 2 psychiatric hospitals.47,48,57–59 The serum cholesterol–lowering diet (high PUFA) consisted of replacing milk with “filled milk,” switching butter for margarine (soft), and liberal use of vegetable oil (mostly soybean oil). The primary outcomes were coronary death and major or intermediate electrocardiograph changes. Serum cholesterol levels were consistently lower in the experimental group, by an average of 13% (P < 0.001). The pooled results of major and intermediate electrocardiograph changes plus CHD death were lower in men (47 in the control vs 25 in the intervention)47 than in women (46 in the control vs 27 in the intervention)48; however, this finding was only statistically significant in men (P < 0.008).47,58 One of the most unique features of the FMHS was the study design. It is the only diet-heart study to utilize a crossover design. The trial was carried out in 2 mental hospitals (N and K). During the first period (1959–1965), Hospital N served as the experimental group, whereas Hospital K served as the control group. The roles were reversed after 6 years. Although referred to as a randomized controlled trial or cluster-randomized trial in multiple publications,11,44,60,61 the FMHS is simply a crossover trial, as noted by the study authors,58 and lacked randomization of the patients. Despite this common misconception, the FMHS was not included or specifically excluded from 8 of the 10 most recent meta-analyses of RCTs for lack of individual randomization, inadequate clusters, inappropriate design to study progressive diseases, and an inability to control for a carryover effect (Table 3).8,10,28,29,45,52 As a consequence of the transient nature of the study population, only 36.4% of the men (246 of 676) and 20.6% of the women (122 of 591) completed both periods of the trial.47,48 Since the cluster design precluded participants acting as their own controls and the study was not randomized, multiple variables were uncontrolled and tended to disadvantage the control group, while benefiting the experimental group. Parodi,43 Ramsden et al,45 and Hamley8 have previously detailed dominant confounders that likely contributed to the results of the FMHS. First, both control groups consumed more trans fat than the intervention groups. Ramsden et al45 determined that Hospital K controls consumed 9 times, and Hospital N 3 times, more TFA than the experimental group. For Hospital K, this amounted to at least 2% more calories from TFA intake compared to the intervention group, which is associated with a 20%–32% increased risk of heart disease event or death.62 In a systematic review and meta-analysis of observational studies, de Souza et al7 found similar results for total trans fat but determined that industrial TFAs – the type used in the FMHS – were independently associated with a 42% increased risk of heart disease. Second, psychotropic and antidepressant medication use was greater in the control groups. The Hospital N control group received 2.18 times more thioridazine. Thioridazine is cardiotoxic and can nearly double the risk of electrocardiograph abnormalities and sudden cardiac death,63,64 which were both endpoints in the study. Of note, the authors stated that electrocardiograph “was one of our principal means of detecting manifestations of CHD” … and that “the most relevant changes in this respect are undoubtedly the Q (including QS) patterns.”47 Specifically, thioridazine has been shown to increase QT-interval prolongation, ventricular arrhythmia, and sudden death to a greater extent than other antipsychotics.65,66 Furthermore, Ramsden et al45 cited evidence that concurrent use of thioridazine with antidepressants can further exacerbate cardiac arrhythmias. Third, in a 1972 Letter to the Editor, Rivers and Yudkin67 noted that the Hospital K control group consumed 49% more sugar per day than the experimental group. Six years later, this disparity disadvantaged the Hospital N control group since they increased their sugar intake by 15%. Although not causal, 10%–24.9% of calories from added sugar have been associated with an increased risk of cardiovascular mortality (HR, 1.30; 95%CI, 1.09–1.55).68 Lastly, the control group had 37 more smokers than the diet group (324 vs 287).47 The authors believed the difference to be minimal; however, the smoking data was never considered during analysis. Furthermore, smoking data was not reported in women participants – only the percentage of nonsmokers.48 Oslo Diet Heart Study The ODHS was conducted with 412 men aged 30–64 years with a previous myocardial infarction from 1956 to 1958. They were randomized 1–2 years later to a control group that consumed their normal diet or an intervention group that consumed a cholesterol-lowering diet low in animal fats and rich in vegetable oil.69 After 5 years of intervention, serum cholesterol was 14% lower in the experimental group and there were significantly fewer major CHD relapses (P = 0.004), but CHD mortality (P = 0.097) and total mortality (P = 0.35) did not reach statistical significance.69,70 When stratified by age, patients with high cholesterol and aged <60 years had greater risk of CHD mortality (P = 0.01), whereas patients aged >60 years showed no association between higher cholesterol and CHD mortality. The success of the ODHS cannot be attributed solely to the replacement of SFA by PUFA or the modification of either, as described by the Advisory.11 According to Hooper et al,28 the ODHS is at high risk for bias owing to an unclear blinding protocol, a systematic difference in care, and, most importantly, a distinct difference in other dietary factors. Blinding is a well-established method for reducing bias in biomedical research. Knowledge can change behavior and blinding attempts to keep the participants, investigators, and outcome assessors unaware of the intervention and/or the outcome of concern. In Leren’s 1966 and 1970 publications,49,70 he acknowledged this lack of blinding and potential for bias, respectively: “In the diagnosis of acquired angina pectoris a personal bias may have been present” and “It should be remembered also that the classification of deaths was not carried out blindly, although it was performed according to precisely defined criteria.” Although speculative, the study author knew the patient assignments and made the determination that participants had experienced angina, the least objective measure, while criteria for sudden death, the most objective measure, remained the same in both groups.70 Second, the experimental group benefited from continuous instruction and supervision by a dietitian, including home visits, letters, and phone calls.70 This systematic difference in care can influence cardiovascular outcomes independent of other changes. A recent systematic review and meta-analysis of 34 studies with 5704 subjects determined that intervention by a registered dietitian compared to usual care improved typical measures of cardiovascular health, including total cholesterol, low-density lipoprotein cholesterol, triglycerides, fasting blood glucose, hemoglobin A1c, and body mass index.71 Lastly, the experimental group was counseled to make multiple dietary changes – to include increasing “salads, beans, peas, cabbage, carrots, fruits, and nuts” and “fish of all types, and all kinds of shell fish … whale beef” – while avoiding the use of sugar.70 The experimental group was given sardines canned in cod-liver oil and soybean oil (15–30 g/d), which significantly increased vitamin D and omega-3 PUFA intake compared to the control group. Based on calculations by Ramsden et al,45 this equates to at least 5 g/d of eicosapentaenoic acid + docosahexaenoic acid, which has been shown to independently reduce post–myocardial infarction sudden death. Most importantly, Leren72 noted that the typical Norwegian diet consisted of about 65 g/d of margarine, and since margarine intake was restricted in the experimental group, Ramsden et al45 determined that the control group consumed up to 9.6% of total energy as TFA. Medical Research Council Study The Medical Research Council study involved 393 men (aged >60 years) recovering from a first myocardial infarction. They were randomly allocated to a normal diet or to an experimental diet characterized by low SFA and high PUFA.54,73 The experimental group were supplemented with 3 oz (85 g) of soybean oil and instructed to consume at least half unheated. The experimental diet reduced cholesterol by 22% after 6 months then slowly increased it over 5 years; however, cholesterol levels were at least 12% below baseline and at least 6% lower than in the control group. Fatal CHD events and major relapses were the same in both groups, but nonmajor relapses were lower in the experimental group (62 vs 74), albeit nonsignificant. Relapses were not associated with initial or follow-up serum cholesterol levels. Similar to FMHS and ODHS, the Medical Research Council defined “forbidden foods” as “butter, other margarines, cooking fat, other oils, fatty meat, whole milk, cheese, egg yolk, and most biscuits and cakes.”54 In practice, the experimental group decreased SFA, while unintentionally reducing TFA intake. Although the experimental group were allowed to have 14 g of “moderately unsaturated margarine” per day,54 the control group most likely consumed more trans fat. Using margarine consumption as a surrogate marker of trans fat, Greaves and Hollingsworth74 noted that margarine intake tripled from 1909 to 1953 (6lb [2.7 kg] vs 18lb [8.2 kg] per person per year) and slowly declined over the next 9 years. The average over this time was 15 lb (6.8 kg) per person per year or 4.6% of total calorie intake per day (average from 1953 to 1962). Since margarine contained 25%–40% TFA,75 this would equate to 1.15%–1.84% of energy as TFA. This value most likely underestimates the total amount of TFA consumed, since shortening, breads, biscuits, and pastries are not included in this total.45 Los Angeles Veterans Administration Trial The LA Veterans trial is only one of 2 double-blind studies to test the diet-heart hypothesis. The trial lasted 8 years and the majority of the patients were enrolled for 6 years. In total, 846 male veterans were randomized to a control (n = 422) or experimental (n = 424) diet.55 The control and experimental groups were fed in separate cafeterias from the rest of the hospital and each other.76 The experimental diet was a high-PUFA diet characterized by replacing animal fats with vegetable oils, and butter with margarine (non-hydrogenated corn oil), using filled milk, and limiting eggs to 1 per day. The participants ranged from 55 to 89 years of age, with an average of 65.5 years. The primary endpoint was coronary artery disease as exhibited by sudden death or acute myocardial infarction. After about 8 years of intervention, serum cholesterol levels were reduced by approximately 13% in the experimental group, but there was no statistically significant difference in the primary endpoints of the study between the groups. However, when these data were pooled with secondary endpoints such as cerebral infarction, ruptured aneurysm, amputation, and others, statistical significance was reached. Deaths due to non-atherosclerotic causes were higher in the experimental group, so total mortality was similar in both groups. As with the other AHA Core Trials, the LA Vets experimental group was instructed to reduce SFA and restrict TFA. Dayton and colleagues77,78 noted that animal fats, hydrogenated shortenings, and conventional margarines were replaced with equal amounts of unsaturated fat, to include un-hydrogenated corn oil–based margarine. Ramsden et al45 estimated this difference in TFA intake to be about 2.1% of total energy. Despite randomization, other dominant confounders were present. For example, the control group included more heavy smokers (n = 25) and fewer nonsmokers (n = 17) and consumed only 16% of the recommended daily allowance for vitamin E compared to 149% in the experimental group, which is especially regrettable, since smoking increases vitamin E requirements79; moreover, withdrawals were more than double in the experimental group, and adherence to the intervention diet was less than 50%.77 In total, major confounding variables – such as concomitant changes in TFA, sugar, omega-3 fatty acids, smoking, and medication use – and potential bias– such as differences in care, adherence, attrition, and lack of blinding – introduced considerable limitations that make interpretation of their results challenging (Table 5).47,49,54,55 Based on the 4CTs alone, it is unclear whether the effects on heart disease were related to replacing SFA with PUFA and undermines the purported strength of this meta-analysis. DISCUSSION This narrative review aimed to provide an overview of the recent meta-analyses of dietary SFA and heart disease and provide a review and critique of the Advisory’s 4 Core Trials leveraged as evidence for the AHA’s dietary SFA recommendations. Meta-analysis, generally recognized as the pinnacle of the hierarchy of evidence,80,81 has become a popular statistical approach to systematically combine clinical trials. From 1995 to 2015, the United States published 16 581 meta-analyses – more than any other country.82 Since 2010, meta-analysis publications have increased by 132%83 and comprise the most cited type of research.84 The meteoric rise in meta-analysis publications has led the scientific community to scrutinize the quality and reliability of meta-analyses. Much of the criticism is related to the fact that meta-analyses evaluate the quantity, not the quality, of observations.83,85,86 A meta-analysis can only be as valid as the studies included in the review. In a 2017 editorial, Dr Milton Packer stated, “In the past, the favored approach was to depict these [patterns] in a narrative, but this task required insight into the details of each trial and a willingness to ask whether differences in design or execution might have contributed to differences in a study’s findings. The current approach to meta-analysis requires no such intellectual effort; little knowledge is needed about any trial, except that it possesses certain minimum features.”85 Fagard et al87 recommended that meta-analyses be approached with caution and meticulously evaluated to determine the risk of publication bias and the potential of malicious influence of inclusion and exclusion criteria. Two recent critiques of nutrition-related meta-analyses determined that individual studies were unreliable, did not meet predefined inclusion criteria, or lacked homogeneity.88,89 Most relevant to the discussion here, Leng90 analyzed dietary fat and heart disease (secondary prevention) review publications from 1969 to 1984. It was determined that 82% (23/28) of reviews that presented supportive views of the diet-heart hypothesis underutilized and selectively cited available RCT evidence. Leng’s analysis confirmed previous findings by Ravnskov.91 In the earliest known review of publication or citation bias in cholesterol-lowering trials, Ravnskov determined that supportive trials were cited 40 times annually, whereas unsupportive trials were cited 7.4 times annually, despite having more trials published in a major journal. To summarize, the Advisory11 seemed to suffer from these same meta-analytical flaws. It only included 4 trials, 3 of the 4 trials did not meet 2 of their 6 inclusion criteria, all of the control/SFA groups in the 4CTs consumed greater TFA than the intervention/PUFA groups, and 3 of the 4CTs contained other major confounding factors (intermittent exposure to treatment diets, smoking, sugar intake, medication, and uncontrolled dietary factors such as vitamin D and omega-3 intake). Interestingly, the Advisory excluded 2 trials—the Minnesota Coronary Survey (MCS) and the Sydney Diet Heart Study (SDHS)—on account of TFA intake in the experimental group, even though 3 of the 4CTs exhibited this same flaw. However, in the Advisory, the disadvantage supported the diet-heart hypothesis. Eight of the 10 most recent meta-analyses included the SDHS, while 7 included the MCS (Table 4).6,8,10,11,22,28,29,44–46 Ramsden et al6,10 recovered data from the excluded trials and found that the SDHS and MCS provided evidence against the traditional diet-heart hypothesis. Despite a decrease in serum cholesterol for both studies (SDHS, 13.3%; MCS, 13.8%), all-cause mortality and CHD mortality were greater in the PUFA group (HR 1.62 and 1.74, respectively)10 or showed no benefit.6 Hamley8 accounted for the Advisory issues in his meta-analysis by differentiating trials as “adequately or inadequately controlled.” His analysis included 5 adequately controlled trials, only 1 of which is included in the 4CTs from the Advisory Medical Research Council. Three of the 4CTs are listed as inadequately controlled for various reasons, including lack of randomization, multifactorial interventions, and lack of control of other variables such as trans fat, vitamin E, and the use of cardiotoxic medication.8 He concluded that adequately controlled trials failed to show a clear association between replacement of SFA with PUFA and CHD events, CHD mortality, or total mortality. The preoccupation with the diet-heart hypothesis has led to an overemphasis of surrogate markers as the primary management tool in reducing heart disease, while neglecting the more meaningful clinical outcome measures such as total mortality, CHD mortality, and CHD events. For example, it is well established that SFA increases LDL-c, while replacing SFA with PUFA reduces LDL-c92,93; however, SFA has not been causally linked to CHD events or mortality and recent evidence has also questioned the benefits of replacing SFA with PUFA (primarily LA).6,8,10 Focusing on surrogate markers has, in turn, caused national agencies to concentrate on the overly simplistic model of single-nutrient recommendations as opposed to recognizing the complex nature of dietary behaviors. To underscore the lack of evidence supporting the link between dietary SFA intake and cardiovascular health, foods containing a high proportion of fat calories as SFA have largely failed to be associated with CVD outcomes (Table 6).94–101 Table 6 Meta-analyses of studies of the effect of foods containing high saturated fat on cardiovascular health Reference . Food product(s) . Main finding(s) . Soedamah-Muthu and de Goede (2018)96 Total dairy Not associated with incidence of CHD Milk Not associated with incidence of CHD and inversely associated with stroke Guo et al (2017)97 High-fat dairy Not associated with mortality, CVD, or CHD Chen et al (2017)98 Cheese Significantly associated with lower risk of CVD Pimpin et al (2016)99 Butter Not significantly associated with any CVD, CHD, or stroke Alexander et al (2016)100 Eggs Not associated with CHD and a reduced risk of stroke O’Connor et al (2017)101 Red meat Does not negatively impact lipoprotein profiles or blood pressure Wang et al (2015)94 Unprocessed red meat Not associated with total mortality, CVD, or CHD Khaw et al (2018)95 Coconut oil Lowered LDL-c compared with butter; no difference in lipids compared with olive oil Reference . Food product(s) . Main finding(s) . Soedamah-Muthu and de Goede (2018)96 Total dairy Not associated with incidence of CHD Milk Not associated with incidence of CHD and inversely associated with stroke Guo et al (2017)97 High-fat dairy Not associated with mortality, CVD, or CHD Chen et al (2017)98 Cheese Significantly associated with lower risk of CVD Pimpin et al (2016)99 Butter Not significantly associated with any CVD, CHD, or stroke Alexander et al (2016)100 Eggs Not associated with CHD and a reduced risk of stroke O’Connor et al (2017)101 Red meat Does not negatively impact lipoprotein profiles or blood pressure Wang et al (2015)94 Unprocessed red meat Not associated with total mortality, CVD, or CHD Khaw et al (2018)95 Coconut oil Lowered LDL-c compared with butter; no difference in lipids compared with olive oil Abbreviations: CHD, coronary heart disease; CVD, cardiovascular disease, LDL-c, low-density lipoprotein cholesterol. Open in new tab Table 6 Meta-analyses of studies of the effect of foods containing high saturated fat on cardiovascular health Reference . Food product(s) . Main finding(s) . Soedamah-Muthu and de Goede (2018)96 Total dairy Not associated with incidence of CHD Milk Not associated with incidence of CHD and inversely associated with stroke Guo et al (2017)97 High-fat dairy Not associated with mortality, CVD, or CHD Chen et al (2017)98 Cheese Significantly associated with lower risk of CVD Pimpin et al (2016)99 Butter Not significantly associated with any CVD, CHD, or stroke Alexander et al (2016)100 Eggs Not associated with CHD and a reduced risk of stroke O’Connor et al (2017)101 Red meat Does not negatively impact lipoprotein profiles or blood pressure Wang et al (2015)94 Unprocessed red meat Not associated with total mortality, CVD, or CHD Khaw et al (2018)95 Coconut oil Lowered LDL-c compared with butter; no difference in lipids compared with olive oil Reference . Food product(s) . Main finding(s) . Soedamah-Muthu and de Goede (2018)96 Total dairy Not associated with incidence of CHD Milk Not associated with incidence of CHD and inversely associated with stroke Guo et al (2017)97 High-fat dairy Not associated with mortality, CVD, or CHD Chen et al (2017)98 Cheese Significantly associated with lower risk of CVD Pimpin et al (2016)99 Butter Not significantly associated with any CVD, CHD, or stroke Alexander et al (2016)100 Eggs Not associated with CHD and a reduced risk of stroke O’Connor et al (2017)101 Red meat Does not negatively impact lipoprotein profiles or blood pressure Wang et al (2015)94 Unprocessed red meat Not associated with total mortality, CVD, or CHD Khaw et al (2018)95 Coconut oil Lowered LDL-c compared with butter; no difference in lipids compared with olive oil Abbreviations: CHD, coronary heart disease; CVD, cardiovascular disease, LDL-c, low-density lipoprotein cholesterol. Open in new tab Despite the established recommendation to eliminate or reduce red meat, full-fat dairy products, butter, and eggs,21 recent meta-analyses have found that total dairy,96 milk,96 high-fat dairy,97 cheese,98 butter,99 eggs,100 and unprocessed red meat94 are not associated with CVD outcomes.96–100 In 2 meta-analyses of RCTs, total red meat intake101,102 was not associated with cardiovascular risk factors. Coconut oil, which is the richest source of SFA at 92% of total fat, decreased LDL-c compared with butter and elicited no change in LDL-c compared with olive oil.95 In the context of a healthy dietary pattern, saturated fat intake does not appear to be harmful. For example, a secondary analysis of a large randomized clinical trial (DIETFITS) found that participants with the greatest increase in the percentage of SFA intake experienced no detrimental effects to their lipid profile while consuming a diet of whole foods and low in refined carbohydrates.103 CONCLUSION This narrative review provides evidence from meta-analyses of observational studies and RCTs that saturated fat intake is not independently associated with the incidence of heart disease and that replacement of SFA with PUFA may not be beneficial despite reducing total- and low-density lipoprotein cholesterol. The clinical trials used by the AHA’s Advisory to uphold their 1961 recommendation to reduce SFA and replace SFA with PUFA are plagued by design flaws, lacked dietary control of variables other than fat, and did not account for trans-fat restriction in the high-PUFA groups. The Ramsden et al6,10 reanalysis of 2 sets of unpublished data, Hamley’s8 meta-analysis of adequately controlled trials, the study selection bias in secondary trials noted by Leng,90 and the Thornley et al53 reanalysis and refutation of the Cochrane meta-analysis strongly challenge the traditional diet-heart hypothesis. The AHA stance regarding the strength of the evidence for the recommendation to limit SFAs to 5%–6% of kcal for those at risk of heart disease may be overstated and in need of reevaluation, especially since some foods with a high SFA profile may not exert LDL-c–elevating effects. The DIETFITS trial should serve as the primary example of the importance of focusing on whole foods instead of single nutrients. Despite the evidence presented, the debate surrounding SFA and heart disease will continue. Whether SFA is a primary driver of heart disease or not, an evidence- and food-based model may be a more meaningful clinical approach to dietary recommendations. As described by Astrup et al,27 recommendations from scientists and researchers to take the complex food matrix into consideration instead of individual nutrients are 9 years old. However, this whole-food approach has only been adopted by a few guideline-producing organizations. The most effective approach may be to provide a simple, straightforward message to the public on whole, unprocessed/minimally processed, nutrient-rich foods that is founded on the best available evidence. Meta-analyses are valuable tools for decision making; however, to determine the best available evidence, individual clinical trials need to be scrutinized to ensure design flaws and bias are adequately considered. Acknowledgments The author thank Dr Yunsuk Koh from the Robbins College of Health and Human Science at Baylor University; Dr Asma S. Bukhari from Walter Reed National Military Medical Center; Dr Julianna Jayne from the Military Nutrition Division, US Army Research Institute of Environmental Medicine; Dr Adam Kieffer from the Army Medical Department Center and School, Health Readiness Center of Excellence; and William Conkright, doctoral student in the Neuromuscular Research Laboratory in the School of Health and Rehabilitation Sciences at the University of Pittsburgh, for encouragement and editorial services. Author contributions. J.L.H. takes sole responsibility for the conceptual framework and content of this manuscript. The views and information presented are those of the author and do not represent the official position of the US Army Medical Center of Excellence, the US Army Training and Doctrine Command, the US Army Medical Command, or the Department of Army, Department of Defense, or US Government. Funding. This research received no external funding. Declaration of interest. The author declares no conflict of interest. References 1 U.S. Department of Health and Human Services and U.S. Department of Agriculture . 2015–2020 Dietary Guidelines for Americans. 8th ed. 2015 . http://health.gov/dietaryguidelines/2015/guidelines/. 2 Eckel RH Jakicic JM Ard JD. 2013 AHA/ACC guideline on lifestyle management to reduce cardiovascular risk: a report of the American College of Cardiology/American Heart Association Task Force on practice guidelines . J Am Coll Cardiol . 2014 ; 63 : 2960 – 2984 . Google Scholar Crossref Search ADS PubMed WorldCat 3 National Health and Medical Research Council . Eat for Health: Australian Dietary Guidelines. Canberra : NHMRC ; 2013 . 4 Kromhout D Spaaij CJ de Goede J , et al. . The 2015 Dutch food-based dietary guidelines . Eur J Clin Nutr. 2016 ; 70 : 869 – 878 . Google Scholar Crossref Search ADS PubMed WorldCat 5 Binns CW Lee MK Kagawa M , et al. . Dietary guidelines for the Asia Pacific region . Asia Pac J Public Health. 2017 ; 29 : 98 – 101 . Google Scholar Crossref Search ADS PubMed WorldCat 6 Ramsden CE Zamora D Majchrzak-Hong S , et al. . Re-evaluation of the traditional diet-heart hypothesis: analysis of recovered data from Minnesota Coronary Experiment (1968-73) . BMJ. 2016 ; 353 : i1246. Google Scholar Crossref Search ADS PubMed WorldCat 7 de Souza RJ Mente A Maroleanu A , et al. . Intake of saturated and trans unsaturated fatty acids and risk of all cause mortality, cardiovascular disease, and type 2 diabetes: systematic review and meta-analysis of observational studies . BMJ . 2015 ; 351 : h3978 . Google Scholar Crossref Search ADS PubMed WorldCat 8 Hamley S. The effect of replacing saturated fat with mostly n-6 polyunsaturated fat on coronary heart disease: a meta-analysis of randomised controlled trials . Nutr J. 2017 ; 16 : 30. Google Scholar Crossref Search ADS PubMed WorldCat 9 Siri-Tarino PW Sun Q Hu FB , et al. . Meta-analysis of prospective cohort studies evaluating the association of saturated fat with cardiovascular disease . Am J Clin Nutr . 2010 ; 91 : 535 – 546 . Google Scholar Crossref Search ADS PubMed WorldCat 10 Ramsden CE Zamora D Leelarthaepin B , et al. . Use of dietary linoleic acid for secondary prevention of coronary heart disease and death: evaluation of recovered data from the Sydney Diet Heart Study and updated meta-analysis . BMJ . 2013 ; 346 : E8707. Google Scholar Crossref Search ADS PubMed WorldCat 11 Sacks FM Lichtenstein AH Wu JHY , et al. . Dietary fats and cardiovascular disease: a presidential advisory from the American Heart Association . Circulation . 2017 ; 136 : e1 – e23 . Google Scholar Crossref Search ADS PubMed WorldCat 12 Liu AG Ford NA Hu FB , et al. . A healthy approach to dietary fats: understanding the science and taking action to reduce consumer confusion . Nutr J. 2017 ; 16 : 53. Google Scholar Crossref Search ADS PubMed WorldCat 13 Anitschkow N Chalatow S. On experimental cholesterin steatosis and its significance in the origin of some pathological processes . Atherosclerosis . 1983 ; 3 : 178 – 182 . OpenURL Placeholder Text WorldCat 14 Messinger WJ Porosowska Y Steele JM. Effect of feeding egg yolk and cholesterol on serum cholesterol levels . Arch Intern Med. 1950 ; 86 : 189 – 195 . Google Scholar Crossref Search ADS WorldCat 15 Gofman JW Lindgren F Elliott H , et al. . The role of lipids and lipoproteins in atherosclerosis . Science . 1950 ; 111 : 166 – 171 . Google Scholar Crossref Search ADS PubMed WorldCat 16 Kinsell LW Partridge J Boling L , et al. . Dietary modification of serum cholesterol and phospholipid levels . J Clin Endocrinol Metab . 1952 ; 7 : 909 – 913 . OpenURL Placeholder Text WorldCat 17 Keys A. Atherosclerosis: a problem in newer public health . J Mt Sinai Hosp N Y. 1953 ; 20 : 118 – 139 . Google Scholar PubMed OpenURL Placeholder Text WorldCat 18 Keys A. The diet and the development of coronary heart disease . J Chron Dis . 1956 ; 4 : 364 – 380 . Google Scholar Crossref Search ADS PubMed WorldCat 19 Keys A. Diet and epidemiology of coronary heart disease . JAMA. 1957 ; 164 : 1912 – 1919 . Google Scholar Crossref Search ADS WorldCat 20 Keys A. Coronary heart disease in seven countries . Circulation . 1970 ; 41 : 118 – 139 . OpenURL Placeholder Text WorldCat 21 Dietary fat and its relation to heart attacks and strokes: report by the Central Committee for Medical and Community Program of the American Heart Association . Circulation . 1961 ; 175 : 133 – 136 . OpenURL Placeholder Text WorldCat 22 Harcombe Z Baker JS DiNicolantonio JJ , et al. . Evidence from randomised controlled trials does not support current dietary fat guidelines: a systematic review and meta-analysis . Open Heart . 2016 ; 3 : e000409. Google Scholar Crossref Search ADS PubMed WorldCat 23 Harcombe Z Baker JS Davies B. Evidence from prospective cohort studies does not support current dietary fat guidelines: a systematic review and meta-analysis . Br J Sports Med. 2017 ; 51 : 1743 – 1749 . Google Scholar Crossref Search ADS PubMed WorldCat 24 Millen BE Abrams S Adams-Campbell L , et al. . The 2015 Dietary Guidelines Advisory Committee scientific report: development and major conclusions . Adv Nutr . 2016 ; 7 : 438 – 444 . Google Scholar Crossref Search ADS PubMed WorldCat 25 Mozaffarian D Ludwig DS. The 2015 US dietary guidelines – lifting the ban on total dietary fat . JAMA . 2015 ; 313 : 2421 – 2422 . Google Scholar Crossref Search ADS PubMed WorldCat 26 Steiber A Tuma PA. Re: Comments on the scientific report of the 2015 Dietary Guidelines Advisory Committee . Off Commun . 2015 :1–15. https://health.gov/dietaryguidelines/dga2015/comments/readCommentDetails.aspx?CID=27125. OpenURL Placeholder Text WorldCat 27 Astrup A Bertram HC Bonjour J-P , et al. . WHO draft guidelines on dietary saturated and trans fatty acids: time for a new approach? BMJ. 2019 ; 366 : l4137. Google Scholar Crossref Search ADS PubMed WorldCat 28 Hooper LS Thompson R Sills D , et al. . Reduced or modified dietary fat for preventing cardiovascular disease . Cochrane Database Syst Rev . 2012 ;5:1–230. OpenURL Placeholder Text WorldCat 29 Hooper L Martin N Abdelhamid A , et al. . G. Reduction in saturated fat intake for cardiovascular disease . Cochrane Database Syst Rev . 2015 ;6:1–158. OpenURL Placeholder Text WorldCat 30 Jakobsen MU O’Reilly EJ Heitmann BL , et al. . Major types of dietary fat and risk of coronary heart disease: a pooled analysis of 11 cohort studies . Am J Clin Nutr . 2009 ; 89 : 1425 – 1432 . Google Scholar Crossref Search ADS PubMed WorldCat 31 Farvid MS Ding M Pan A , et al. . Dietary linoleic acid and risk of coronary heart disease: a systematic review and meta-analysis of prospective cohort studies . Circulation . 2014 ; 130 : 1568 – 1578 . Google Scholar Crossref Search ADS PubMed WorldCat 32 Chowdhury R Warnakula S Kunutsor S , et al. . Association of dietary, circulating, and supplement fatty acids with coronary risk: a systematic review and meta-analysis . Ann Intern Med. 2014 ; 160 : 398 – 406 . Google Scholar Crossref Search ADS PubMed WorldCat 33 Skeaff CM Miller J. Dietary fat and coronary heart disease: summary of evidence from prospective cohort and randomised controlled trials . Ann Nutr Metab. 2009 ; 55 : 173 – 201 . Google Scholar Crossref Search ADS PubMed WorldCat 34 Mente A de Koning L Shannon HS , et al. . A systematic review of the evidence supporting a causal link between dietary factors and coronary heart disease . Arch Intern Med. 2009 ; 169 : 659 – 669 . Google Scholar Crossref Search ADS PubMed WorldCat 35 Zhu Y Bo Y Liu Y. Dietary total fat, fatty acids intake, and risk of cardiovascular disease: a dose-response meta-analysis of cohort studies . Lipids Health Dis. 2019 ; 18 : 91. Google Scholar Crossref Search ADS PubMed WorldCat 36 Kris-Etherton P Fleming J Harris WS. Dietary omega-6 polyunsaturated fatty acids – important for heart health . Clin Nutr Insight . 2009 ; 35 : 1 – 5 . Google Scholar Crossref Search ADS WorldCat 37 Katan MB. Omega-6 polyunsaturated fatty acids and coronary heart disease . Am J Clin Nutr . 2009 ; 89 : 1283 – 1284 . Google Scholar Crossref Search ADS PubMed WorldCat 38 Kris-Etherton P Fleming J Harris WS. The debate about n-6 polyunsaturated fatty acid recommendations for cardiovascular health . J Am Diet Assoc . 2010 ; 110 : 201 – 204 . Google Scholar Crossref Search ADS PubMed WorldCat 39 Jayedi A Shab-Bidar S Eimeri S , et al. . Fish consumption and risk of all-cause and cardiovascular mortality: a dose-response meta-analysis of prospective observational studies . Public Health Nutr. 2018 ; 21 : 1297 – 1306 . Google Scholar Crossref Search ADS PubMed WorldCat 40 Jayedi A Zargar MS Shab-Bidar S. Fish consumption and risk of myocardial infarction: a systematic review and dose-response meta-analysis suggests a regional difference . Nutr Res . 2019 ; 62 : 1 – 12 . Google Scholar Crossref Search ADS PubMed WorldCat 41 Song M Giovannucci E. Substitution analysis in nutritional epidemiology: proceed with caution . Eur J Epidemiol. 2018 ; 33 : 137 – 140 . Google Scholar Crossref Search ADS PubMed WorldCat 42 Rehm CD Drewnowski A. Replacing dairy fat with polyunsaturated and monounsaturated fatty acids: a food-level modeling study of dietary nutrient density and diet quality using the 2013-16 National Health and Nutrition Examination Survey . Front Nutr . 2019 ; 6 : 113 . Google Scholar Crossref Search ADS PubMed WorldCat 43 Parodi PW. Has the association between saturated fatty acids, serum cholesterol and coronary heart disease been over emphasized? Int Dairy J . 2009 ; 19 : 345 – 361 . Google Scholar Crossref Search ADS WorldCat 44 Mozaffarian D Micha R Wallace S. Effects on coronary heart disease of increasing polyunsaturated fat in place of saturated fat: a systematic review and meta-analysis of randomized controlled trials . PLoS Med. 2010 ; 7 : e1000252. Google Scholar Crossref Search ADS PubMed WorldCat 45 Ramsden CE Hibbeln JR Majchrzak SF , et al. . n-6 fatty acid-specific and mixed polyunsaturate dietary interventions have different effects on CHD risk: a meta-analysis of randomised controlled trials . Br J Nutr. 2010 ; 104 : 1586 – 1600 . Google Scholar Crossref Search ADS PubMed WorldCat 46 Schwingshackl L Hoffmann G. Dietary fatty acids in the secondary prevention of coronary heart disease: a systematic review, meta-analysis and meta-regression . BMJ Open . 2014 ; 4 : e004487. Google Scholar Crossref Search ADS PubMed WorldCat 47 Turpeinen O Karvonen MJ Pekkarinen M , et al. . Dietary prevention of coronary heart disease: the Finnish Mental Hospital Study . Int J Epidemiol. 1979 ; 8 : 99 – 118 . Google Scholar Crossref Search ADS PubMed WorldCat 48 Miettinen M Turpeinen O Karvonen MJ , et al. . Dietary prevention of coronary heart disease in women: the Finnish Mental Hospital Study . Int J Epidemiol . 1983 ; 12 : 17 – 25 . Google Scholar Crossref Search ADS PubMed WorldCat 49 Leren P. The Oslo diet heart study: eleven-year report . Circulation . 1970 ; 42 : 935 – 942 . Google Scholar Crossref Search ADS PubMed WorldCat 50 Rose GA Thomson WB Williams RT. Corn oil in treatment of ischaemic heart disease . Br Med J . 1965 ; 1 : 1531 – 1533 . Google Scholar Crossref Search ADS PubMed WorldCat 51 Hooper L Summerbell CD Higgins JP , et al. . Reduced or modified dietary fat for preventing cardiovascular disease . Cochrane Database Syst Rev . 2001 ;3:1–56. OpenURL Placeholder Text WorldCat 52 Harcombe Z Baker JS Cooper SM , et al. . Evidence from randomised controlled trials did not support the introduction of dietary fat guidelines in 1977 and 1983: a systematic review and meta-analysis . Open Heart . 2015 ; 2 : e000196. Google Scholar Crossref Search ADS PubMed WorldCat 53 Thornley S Schofield G Zinn C , et al. . How reliable is the statistical evidence for limiting saturated fat intake? A fresh look at the influential Hooper meta-analysis . Intern Med J. 2019 . doi:10.1111/imj.14325 OpenURL Placeholder Text WorldCat 54 MRC . Controlled trial of soya-bean oil in myocardial infarction . Lancet . 1968 ; 28 : 693 – 699 . OpenURL Placeholder Text WorldCat 55 Dayton S Pearce ML. Prevention of coronary heart disease and other complications of atherosclerosis by modified diet . Am J Med . 1969 ; 46 : 751 – 762 . Google Scholar Crossref Search ADS PubMed WorldCat 56 Knopp RH Retzlaff BM. Saturated fat prevents coronary artery disease? An American paradox . Am J Clin Nutr . 2004 ; 80 : 1102 – 1103 . Google Scholar Crossref Search ADS PubMed WorldCat 57 Turpeinen O Miettinen M Karvonen MJ , et al. . Dietary prevention of coronary heart disease: long-term experiment. I. Observations on male subjects . Am J Clin Nutr . 1968 ; 21 : 255 – 276 . Google Scholar Crossref Search ADS PubMed WorldCat 58 Miettinen M Turpeinen O Karvonen MJ , et al. . Effect of cholesterol-lowering diet on mortality from coronary heart-disease and other causes. A twelve-year clinical trial in men and women . Lancet (London, England) . 1972 ; 2 : 835 – 838 . Google Scholar Crossref Search ADS PubMed WorldCat 59 Turpeinen O. Effect of cholesterol-lowering diet on mortality from coronary heart disease and other causes . Circulation . 1979 ; 59 : 1 – 7 . Google Scholar Crossref Search ADS PubMed WorldCat 60 Siri-Tarino PW Sun Q Hu FB , et al. . Saturated fat, carbohydrate, and cardiovascular disease . Am J Clin Nutr . 2010 ; 91 : 502 – 509 . Google Scholar Crossref Search ADS PubMed WorldCat 61 Wang DD Hu FB. Dietary fat and risk of cardiovascular disease: recent controversies and advances . Annu Rev Nutr. 2017 ; 37 : 423 – 446 . Google Scholar Crossref Search ADS PubMed WorldCat 62 Mozaffarian D Aro A Willett WC. Health effects of trans-fatty acids: experimental and observational evidence . Eur J Clin Nutr. 2009 ; 63(suppl 2) : S5 – S21 . Google Scholar Crossref Search ADS PubMed WorldCat 63 Sicouri S Antzelevitch C. Sudden cardiac death secondary to antidepressant and antipsychotic drugs . Expert Opin Drug Saf . 2008 ; 7 : 181 – 194 . Google Scholar Crossref Search ADS PubMed WorldCat 64 Liperoti R Gambassi G Lapane KL , et al. . Conventional and atypical antipsychotics and the risk of hospitalization for ventricular arrhythmia or cardiac arrest . Arch Intern Med. 2005 ; 165 : 696 – 701 . Google Scholar Crossref Search ADS PubMed WorldCat 65 Abdelmawla N Mitchell A. Sudden cardiac death and antipsychotics. Part 1: risk factors and mechanisms . Adv Psychiatr Treat. 2006 ; 12 : 35 – 44 . Google Scholar Crossref Search ADS WorldCat 66 Glassman AH Bigger JT. Antipsychotic drugs prolonged QTc interval, TdP, and sudden death . Am J Psychiatry . 2001 ; 158 : 1774 – 1782 . Google Scholar Crossref Search ADS PubMed WorldCat 67 Rivers J Yudkin J. Cholesterol-lowering diet and mortality from coronary heart-disease . Lancet . 1972 ; 2 : 1026 – 1027 . Google Scholar Crossref Search ADS PubMed WorldCat 68 Yang Q Zhang Z Gregg EW , et al. . Added sugar intake and cardiovascular diseases mortality among US adults . JAMA Intern Med. 2014 ; 174 : 516 – 524 . Google Scholar Crossref Search ADS PubMed WorldCat 69 Leren P. The effect of plasma-cholesterol-lowering diet in male survivors of myocardial infarction: a controlled clinical trial . Bull N Y Acad Med . 1968 ; 44 : 1012 – 1020 . Google Scholar PubMed OpenURL Placeholder Text WorldCat 70 Leren P. The effect of plasma cholesterol lowering diet in male survivors of myocardial infarction: a controlled clinical trial . Acta Med Scand Suppl. 1966 ; 466 : 1 – 92 . Google Scholar PubMed OpenURL Placeholder Text WorldCat 71 Sikand G Cole RE Handu D , et al. . Clinical and cost benefits of medical nutrition therapy by registered dietitian nutritionists for management of dyslipidemia: a systematic review and meta-analysis . J Clin Lipidol . 2018 ; 12 : 1113 – 1122 . Google Scholar Crossref Search ADS PubMed WorldCat 72 Leren P. Prevention of coronary heart disease: some results from the Oslo secondary and primary intervention studies . J Am Coll Nutr . 1989 ; 8 : 407 – 410 . Google Scholar Crossref Search ADS PubMed WorldCat 73 Clarke JAC Hedley E Marr JW , et al. . Dietary aspects of a controlled trial of soya-bean oil in myocardial infarction . Int J Food Sci Nutr. 1969 ; 23 : 136 – 150 . Google Scholar Crossref Search ADS WorldCat 74 Greaves JP Hollingsworth DF. Trends in food consumption in the United Kingdom . World Rev Nutr Diet. 1966 ; 6 : 34 – 89 . Google Scholar Crossref Search ADS PubMed WorldCat 75 Brown JB. Changes in nutritive value of food fats during processing and cooking . Nutr Rev . 2009 ; 17 : 321 – 325 . Google Scholar Crossref Search ADS WorldCat 76 Los Angeles Veterans Administration diet study . Nutr Rev . 1969 ; 27 : 311 – 316 . PubMed OpenURL Placeholder Text WorldCat 77 Dayton S Pearce ML Hashimoto S , et al. . A controlled clinical trial of a diet high in unsaturated fat in preventing complication of atherosclerosis . Circulation . 1969 ; 1 : 1 – 63 . OpenURL Placeholder Text WorldCat 78 Hiscock E Dayton S Pearce ML , et al. . A palatable diet high in unsaturated fat . J Am Diet Assoc. 1962 ; 40 : 427 – 431 . Google Scholar PubMed OpenURL Placeholder Text WorldCat 79 Bruno RS Traber MG. Cigarette smoke alters human vitamin E requirements . J Nutr . 2005 ; 135 : 671 – 674 . Google Scholar Crossref Search ADS PubMed WorldCat 80 Hierarchy of evidence and grading of recommendations . Thorax . 2004 ; 59(suppl 1) : i13 – i14 . OpenURL Placeholder Text WorldCat 81 Evans D. Hierarchy of evidence: a framework for ranking evidence evaluating healthcare interventions . J Clin Nurs. 2003 ; 12 : 77 – 84 . Google Scholar Crossref Search ADS PubMed WorldCat 82 Fontelo P Liu F. A review of recent publication trends from top publishing countries . Syst Rev. 2018 ; 7 : 147. Google Scholar Crossref Search ADS PubMed WorldCat 83 Ioannidis JP. The mass production of redundant, misleading, and conflicted systematic reviews and meta-analyses . Milbank Q . 2016 ; 94 : 485 – 514 . Google Scholar Crossref Search ADS PubMed WorldCat 84 Patsopoulos NA Analatos AA Ioannidis J. Relative citation impact of various study designs in the health sciences . JAMA . 2005 ; 293 : 2362 – 2366 . Google Scholar Crossref Search ADS PubMed WorldCat 85 Packer M. Are meta-analyses a form of medical fake news? Thoughts about how they should contribute to medical science and practice . Circulation . 2017 ; 136 : 2097 – 2099 . Google Scholar Crossref Search ADS PubMed WorldCat 86 Barnard ND Willett WC Ding EL. The misuse of meta-analysis in nutrition research . JAMA . 2017 ; 318 : 1435 – 1436 . Google Scholar Crossref Search ADS PubMed WorldCat 87 Fagard RH Staessen JA Thijs L. Advantages and disadvantages of the meta-analysis approach . J Hypertens Suppl . 1996 ; 14 : S9 ; discussion S13. Google Scholar Crossref Search ADS PubMed WorldCat 88 Peace KE Yin J Rochani H , et al. . A serious flaw in nutrition epidemiology: a meta-analysis study . Int J Biostat . 2018 ; 14 :1–10. OpenURL Placeholder Text WorldCat 89 Tabung FK. Inaccurate data in meta-analysis ‘Dietary patterns and colorectal cancer risk: a meta-analysis’ . Eur J Cancer Prev . 2019 ; 28 : 58. Google Scholar Crossref Search ADS PubMed WorldCat 90 Leng RI. A network analysis of the propagation of evidence regarding the effectiveness of fat-controlled diets in the secondary prevention of coronary heart disease (CHD): selective citation in reviews . PLoS One. 2018 ; 13 : e0197716 . Google Scholar Crossref Search ADS PubMed WorldCat 91 Ravnskov U. Cholesterol lowering trials in coronary heart disease: frequency of citation and outcome . BMJ . 1992 ; 305 : 15 – 19 . Google Scholar Crossref Search ADS PubMed WorldCat 92 Mensink RP Zock PL Kester ADM , et al. . Effects of dietary fatty acids and carbohydrates on the ratio of serum total to HDL cholesterol and on serum lipids and apolipoproteins: a meta-analysis of 60 controlled trials . Am J Clin Nutr . 2003 ; 77 : 1146 – 1155 . Google Scholar Crossref Search ADS PubMed WorldCat 93 Mensink RP. Effects of saturated fatty acids on serum lipids and lipoproteins: a systematic review and regression analysis . World Health Organization; 2016 :1–63. OpenURL Placeholder Text WorldCat 94 Wang X Lin X Ouyang YY , et al. . Red and processed meat consumption and mortality: dose–response meta-analysis of prospective cohort studies . Public Health Nutr. 2016 ; 19 : 893 – 905 . Google Scholar Crossref Search ADS PubMed WorldCat 95 Khaw KT Sharp SJ Finikarides L , et al. . Randomised trial of coconut oil, olive oil or butter on blood lipids and other cardiovascular risk factors in healthy men and women . BMJ Open . 2018 ; 8 : e020167. Google Scholar Crossref Search ADS PubMed WorldCat 96 Soedamah-Muthu SS de Goede J. Dairy consumption and cardiometabolic diseases: systematic review and updated meta-analyses of prospective cohort studies . Curr Nutr Rep. 2018 ; 7 : 171 – 182 . Google Scholar Crossref Search ADS PubMed WorldCat 97 Guo J Astrup A Lovegrove JA , et al. . Milk and dairy consumption and risk of cardiovascular diseases and all-cause mortality: dose-response meta-analysis of prospective cohort studies . Eur J Epidemiol. 2017 ; 32 : 269 – 287 . Google Scholar Crossref Search ADS PubMed WorldCat 98 Chen GC Wang Y Tong X , et al. . Cheese consumption and risk of cardiovascular disease: a meta-analysis of prospective studies . Eur J Nutr. 2017 ; 56 : 2565 – 2575 . Google Scholar Crossref Search ADS PubMed WorldCat 99 Pimpin L Wu JH Haskelberg H , et al. . Is butter back? A systematic review and meta-analysis of butter consumption and risk of cardiovascular disease, diabetes, and total mortality . PLoS One. 2016 ; 11 : e0158118 . Google Scholar Crossref Search ADS PubMed WorldCat 100 Alexander DD Miller PE Vargas AJ , et al. . Meta-analysis of egg consumption and risk of coronary heart disease and stroke . J Am Coll Nutr . 2016 ; 35 : 704 – 716 . Google Scholar Crossref Search ADS PubMed WorldCat 101 O'Connor LE Kim JE Campbell WW. Total red meat intake of >/=0.5 servings/d does not negatively influence cardiovascular disease risk factors: a systemically searched meta-analysis of randomized controlled trials . Am J Clin Nutr. 2017 ; 105 : 57 – 69 . Google Scholar Crossref Search ADS PubMed WorldCat 102 Guasch-Ferre M Satija A Blondin SA , et al. . Meta-analysis of randomized controlled trials of red meat consumption in comparison with various comparison diets on cardiovascular risk factors . Circulation . 2019 ; 139 : 1828 – 1845 . Google Scholar Crossref Search ADS PubMed WorldCat 103 Shih CW Hauser ME Aronica L , et al. . Changes in blood lipid concentrations associated with changes in intake of dietary saturated fat in the context of a healthy low-carbohydrate weight-loss diet: a secondary analysis of the Diet Intervention Examining The Factors Interacting with Treatment Success (DIETFITS) trial . Am J Clin Nutr . 2019 ; 109 : 433 – 441 . Google Scholar Crossref Search ADS PubMed WorldCat © The Author(s) 2019. Published by Oxford University Press on behalf of the International Life Sciences Institute. All rights reserved. For permissions, please e-mail: [email protected]. This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/open_access/funder_policies/chorus/standard_publication_model)

Coelho, Mariana O, C;Monteyne, Alistair, J;Dunlop, Mandy, V;Harris, Hannah, C;Morrison, Douglas, J;Stephens, Francis, B;Wall, Benjamin, T

doi: 10.1093/nutrit/nuz077pmid: 31841152

Abstract The world’s population is expanding, leading to an increased global requirement for dietary protein to support health and adaptation in various populations. Though a strong evidence base supports the nutritional value of animal-derived dietary proteins, mounting challenges associated with sustainability of these proteins have led to calls for the investigation of alternative, non–animal-derived dietary protein sources. Mycoprotein is a sustainably produced, protein-rich, high-fiber, whole food source derived from the fermentation of fungus. Initial investigations in humans demonstrated that mycoprotein consumption can lower circulating cholesterol concentrations. Recent data also report improved acute postprandial glycemic control and a potent satiety effect following mycoprotein ingestion. It is possible that these beneficial effects are attributable to the amount and type of dietary fiber present in mycoprotein. Emerging data suggest that the amino acid composition and bioavailability of mycoprotein may also position it as a promising dietary protein source to support skeletal muscle protein metabolism. Mycoprotein may be a viable dietary protein source to promote training adaptations in athletes and the maintenance of muscle mass to support healthy aging. Herein, current evidence underlying the metabolic effects of mycoprotein is reviewed, and the key questions to be addressed are highlighted. cardiometabolic health, cholesterol, muscle protein metabolism, mycoprotein, sarcopenia, weight management INTRODUCTION Developing a nutritionally sustainable future for the world’s population, projected to grow from approximately 7.3 billion to more than 9 billion by 2050,1 is an urgent contemporary issue. This projected increase is coupled with global trends of rises in urbanization, social mobility, and wealth creation, all factors expected to exacerbate the global food demand.1 As a result, current and future generations must view developments in the understanding of human nutrition through the lens of mounting challenges associated with the sustainability of increased food production. The demand to produce sufficient dietary protein for the world’s population is driven by demographics and compounded by accumulating scientific data that support protein consumption at levels greater than the currently accepted Recommended Dietary Allowances in various populations. For instance, evidence suggests that muscle mass maintenance in older adults,2–5 the promotion or retention of muscular training adaptations in athletes,6,7 and successful weight management8 are all supported by modest increases in dietary protein intake above the currently accepted Recommended Dietary Allowances. It is clear, therefore, that the global need for dietary protein production is a pressing societal issue that is gathering momentum. The majority of data supporting the refinement of dietary protein requirements has been obtained from studies examining the in vivo metabolic handling or adaptive responses to animal-derived protein ingestion.9–11 The carbon, water, and land-use footprints of animal-derived protein production are 8 to 80 times, 50 to 150 times, and 30 to 220 times greater, respectively, than those of many plant-based proteins (with variations dependent on the protein source and the method of quantification).12 Furthermore, vegan, vegetarian, and flexitarian diets are increasing in popularity.13 As a result, research that investigates non–animal-derived protein sources is applicable to a progressively larger demographic, and the impact of such evidence will increase correspondingly. It is therefore vital that the scientific community begin to examine the metabolic handling and nutritional value of alternative, non–animal-derived, sustainable sources of protein. MYCOPROTEIN Mycoprotein is a whole food source produced by continuous flow fermentation of the filamentous fungus Fusarium venenatum (a detailed description of the production processes is provided by Finnigan14). The resultant product is a high-protein, high-fiber, and relatively low-energy complete food source (Table 115–17 and Table 215,16) that is textured (via freezing) and flavored into a variety of products under the trade name Quorn (Marlow Foods; Stokesley, North Yorkshire, UK). The sustainability credentials of mycoprotein production position this protein as an attractive alternative protein source to temper environmental concerns associated with increased production of dietary protein18,19 (Figure 1). Figure 1 Open in new tabDownload slide Greenhouse gas emissions and water use required to produce a 30-g portion of protein from beef mince, milk, chicken, Quorn mince, Quorn pieces, and mycoprotein. Data were taken from Quorn, Beef and Chicken Footprints,19 with additional data provided by (and reproduced with permission of) the Carbon Trust, London, UK. Abbreviations: kgCO2e, kilograms carbon dioxide equivalent; GHG, greenhouse gas emissions. Figure 1 Open in new tabDownload slide Greenhouse gas emissions and water use required to produce a 30-g portion of protein from beef mince, milk, chicken, Quorn mince, Quorn pieces, and mycoprotein. Data were taken from Quorn, Beef and Chicken Footprints,19 with additional data provided by (and reproduced with permission of) the Carbon Trust, London, UK. Abbreviations: kgCO2e, kilograms carbon dioxide equivalent; GHG, greenhouse gas emissions. Table 1 Nutritional content of mycoprotein, commercially available protein isolates, Quorn vegan pieces, and a selection of commonly consumed protein sourcesa Protein source . Nutrient composition per 100 g . Leucine matchedb . Protein (g) . Fat (g) . Carbohydrate (g) . Fiber (g) . Energy (kcal) . Energy (kJ) . Leucine (g) . Product (g) . Protein (g) . Mycoprotein (dry weight) 45 13 10 25 340 1423 3.9 64 29  Whey protein 80 7 5 < 0.1 402 1682 8.6 29 23  Milk protein 80 1 6 < 0.1 350 1464 7.0 36 29  Quorn pieces 15 3 4 5 113 473 1.2 208 32  Whole egg, raw 13 10 < 1 <1 143 598 1.1 230 29 Beef mince (5%), raw 21 5 0 0 137 573 1.7 150 32  Chicken meat, raw 21 3 0 0 119 498 1.6 156 33  Cod meat, raw 18 1 0 0 82 343 1.4 173 31 Protein source . Nutrient composition per 100 g . Leucine matchedb . Protein (g) . Fat (g) . Carbohydrate (g) . Fiber (g) . Energy (kcal) . Energy (kJ) . Leucine (g) . Product (g) . Protein (g) . Mycoprotein (dry weight) 45 13 10 25 340 1423 3.9 64 29  Whey protein 80 7 5 < 0.1 402 1682 8.6 29 23  Milk protein 80 1 6 < 0.1 350 1464 7.0 36 29  Quorn pieces 15 3 4 5 113 473 1.2 208 32  Whole egg, raw 13 10 < 1 <1 143 598 1.1 230 29 Beef mince (5%), raw 21 5 0 0 137 573 1.7 150 32  Chicken meat, raw 21 3 0 0 119 498 1.6 156 33  Cod meat, raw 18 1 0 0 82 343 1.4 173 31 a Data adapted from the internal analyses of Dunlop et al (2017),15 from Gorissen et al (2018),16 and from Food Data Central.17 Values are approximations, based on the data available. b Reflects the approximate amount of product and protein required to be consumed to obtain 2.5 g of leucine. Open in new tab Table 1 Nutritional content of mycoprotein, commercially available protein isolates, Quorn vegan pieces, and a selection of commonly consumed protein sourcesa Protein source . Nutrient composition per 100 g . Leucine matchedb . Protein (g) . Fat (g) . Carbohydrate (g) . Fiber (g) . Energy (kcal) . Energy (kJ) . Leucine (g) . Product (g) . Protein (g) . Mycoprotein (dry weight) 45 13 10 25 340 1423 3.9 64 29  Whey protein 80 7 5 < 0.1 402 1682 8.6 29 23  Milk protein 80 1 6 < 0.1 350 1464 7.0 36 29  Quorn pieces 15 3 4 5 113 473 1.2 208 32  Whole egg, raw 13 10 < 1 <1 143 598 1.1 230 29 Beef mince (5%), raw 21 5 0 0 137 573 1.7 150 32  Chicken meat, raw 21 3 0 0 119 498 1.6 156 33  Cod meat, raw 18 1 0 0 82 343 1.4 173 31 Protein source . Nutrient composition per 100 g . Leucine matchedb . Protein (g) . Fat (g) . Carbohydrate (g) . Fiber (g) . Energy (kcal) . Energy (kJ) . Leucine (g) . Product (g) . Protein (g) . Mycoprotein (dry weight) 45 13 10 25 340 1423 3.9 64 29  Whey protein 80 7 5 < 0.1 402 1682 8.6 29 23  Milk protein 80 1 6 < 0.1 350 1464 7.0 36 29  Quorn pieces 15 3 4 5 113 473 1.2 208 32  Whole egg, raw 13 10 < 1 <1 143 598 1.1 230 29 Beef mince (5%), raw 21 5 0 0 137 573 1.7 150 32  Chicken meat, raw 21 3 0 0 119 498 1.6 156 33  Cod meat, raw 18 1 0 0 82 343 1.4 173 31 a Data adapted from the internal analyses of Dunlop et al (2017),15 from Gorissen et al (2018),16 and from Food Data Central.17 Values are approximations, based on the data available. b Reflects the approximate amount of product and protein required to be consumed to obtain 2.5 g of leucine. Open in new tab Table 2 Amino acid content of mycoprotein and commercially available protein isolatesa Amino acid . Grams per 100 g of mycoprotein (dw) . Grams per 100 g of whey protein . Grams per 100 g of milk protein . Grams per 100 g of egg protein . Alanine 2.8 4.2 2.6 2.6 Arginine 3.3 1.7 2.6 2.6 Aspartic acid 4.6 Cystine 0.4 0.8 0.2 0.4 Glutamic acid 5.6 15.5 16.7 5.1 Glycine 2.0 1.5 1.5 1.4 Histidine 1.6 1.4 1.9 0.9 Isoleucine 2.4 3.8 2.9 1.6 Leucine 3.9 8.6 7 3.6 Lysine 3.8 7.1 5.9 2.7 Methionine 1.0 1.8 2.1 1.4 Phenylalanine 2.3 2.5 3.5 2.3 Proline 2.0 4.8 7.3 1.8 Serine 2.3 4 4 3.3 Threonine 2.5 5.4 3.5 2 Tryptophan 0.8 Tyrosine 1.8 2.4 3.8 1.8 Valine 2.8 3.5 3.6 2 EAAs 20.9 34.1 30.4 16.5 NEAAs 24.6 34.9 38.7 19.0 BCAAs 9.0 15.9 13.5 7.2 Amino acid . Grams per 100 g of mycoprotein (dw) . Grams per 100 g of whey protein . Grams per 100 g of milk protein . Grams per 100 g of egg protein . Alanine 2.8 4.2 2.6 2.6 Arginine 3.3 1.7 2.6 2.6 Aspartic acid 4.6 Cystine 0.4 0.8 0.2 0.4 Glutamic acid 5.6 15.5 16.7 5.1 Glycine 2.0 1.5 1.5 1.4 Histidine 1.6 1.4 1.9 0.9 Isoleucine 2.4 3.8 2.9 1.6 Leucine 3.9 8.6 7 3.6 Lysine 3.8 7.1 5.9 2.7 Methionine 1.0 1.8 2.1 1.4 Phenylalanine 2.3 2.5 3.5 2.3 Proline 2.0 4.8 7.3 1.8 Serine 2.3 4 4 3.3 Threonine 2.5 5.4 3.5 2 Tryptophan 0.8 Tyrosine 1.8 2.4 3.8 1.8 Valine 2.8 3.5 3.6 2 EAAs 20.9 34.1 30.4 16.5 NEAAs 24.6 34.9 38.7 19.0 BCAAs 9.0 15.9 13.5 7.2 Abbreviations: BCAAs, branched-chain amino acids; dw, dry weight; EAAs, essential amino acids; NEAAs, nonessential amino acids. a Data adapted from the internal analyses of Dunlop et al (2017)15 and from Gorissen et al (2018).16 Open in new tab Table 2 Amino acid content of mycoprotein and commercially available protein isolatesa Amino acid . Grams per 100 g of mycoprotein (dw) . Grams per 100 g of whey protein . Grams per 100 g of milk protein . Grams per 100 g of egg protein . Alanine 2.8 4.2 2.6 2.6 Arginine 3.3 1.7 2.6 2.6 Aspartic acid 4.6 Cystine 0.4 0.8 0.2 0.4 Glutamic acid 5.6 15.5 16.7 5.1 Glycine 2.0 1.5 1.5 1.4 Histidine 1.6 1.4 1.9 0.9 Isoleucine 2.4 3.8 2.9 1.6 Leucine 3.9 8.6 7 3.6 Lysine 3.8 7.1 5.9 2.7 Methionine 1.0 1.8 2.1 1.4 Phenylalanine 2.3 2.5 3.5 2.3 Proline 2.0 4.8 7.3 1.8 Serine 2.3 4 4 3.3 Threonine 2.5 5.4 3.5 2 Tryptophan 0.8 Tyrosine 1.8 2.4 3.8 1.8 Valine 2.8 3.5 3.6 2 EAAs 20.9 34.1 30.4 16.5 NEAAs 24.6 34.9 38.7 19.0 BCAAs 9.0 15.9 13.5 7.2 Amino acid . Grams per 100 g of mycoprotein (dw) . Grams per 100 g of whey protein . Grams per 100 g of milk protein . Grams per 100 g of egg protein . Alanine 2.8 4.2 2.6 2.6 Arginine 3.3 1.7 2.6 2.6 Aspartic acid 4.6 Cystine 0.4 0.8 0.2 0.4 Glutamic acid 5.6 15.5 16.7 5.1 Glycine 2.0 1.5 1.5 1.4 Histidine 1.6 1.4 1.9 0.9 Isoleucine 2.4 3.8 2.9 1.6 Leucine 3.9 8.6 7 3.6 Lysine 3.8 7.1 5.9 2.7 Methionine 1.0 1.8 2.1 1.4 Phenylalanine 2.3 2.5 3.5 2.3 Proline 2.0 4.8 7.3 1.8 Serine 2.3 4 4 3.3 Threonine 2.5 5.4 3.5 2 Tryptophan 0.8 Tyrosine 1.8 2.4 3.8 1.8 Valine 2.8 3.5 3.6 2 EAAs 20.9 34.1 30.4 16.5 NEAAs 24.6 34.9 38.7 19.0 BCAAs 9.0 15.9 13.5 7.2 Abbreviations: BCAAs, branched-chain amino acids; dw, dry weight; EAAs, essential amino acids; NEAAs, nonessential amino acids. a Data adapted from the internal analyses of Dunlop et al (2017)15 and from Gorissen et al (2018).16 Open in new tab Following the development of mycoprotein in the 1960s, initial experiments in humans during the 1970s established the basic feasibility, tolerability, and metabolic impact of mycoprotein consumption, after which mycoprotein became commercially available in 1985. By the end of the 1990s, this research in humans had begun to wane. A complete list and summary of published human mycoprotein studies performed to date is shown in Table 3.15,20–31 In recent years, however, the established commercial viability, environmental advantages, and alternative potential applications of mycoprotein for metabolic health, as well as for skeletal muscle maintenance and reconditioning, have reignited interest in this novel food source. Table 3 Human studies investigating the metabolic effects of mycoprotein Reference . No. of participants . Participants . Type of study . Type of intervention . Duration of intervention . Study findings . Udall et al (1984)20 100 Healthy adults Double-blind crossover trial Mycoprotein-based cookie supplementation (20 g dw per day) vs control cookies 30 d 6.9% ↓ in PC No changes in BW or other blood markers (glucose, urea, nitrogen, sodium, potassium, calcium, phosphorus, uric acid, creatinine, lactic acid dehydrogenase, alkaline phosphatase, amylase, SGOT, total protein, albumin, TGs, CBC) No changes in urine markers (pH, glucose, protein, ketones, white and red blood cells) Turnbull et al (1990)21 17 (9 mycoprotein, 8 control) Healthy adults with total cholesterol between 5.2 and 6.2 mmol/L Randomized controlled parallel group trial Mycoprotein (≈ 191 g Quorn per day) vs meat during a fully controlled diet 3 wk 13% ↓ in PC 9% ↓ in plasma LDL-C (12% ↑ in control group) 12% ↑ in plasma HDL-C (11% ↓ in control group) 53% ↓ in TGs (in both groups) No differences in BW or BP No changes in fasting insulin and glucose No changes in Apo A-I or Apo B Turnbull et al (1992)22 21 (11 mycoprotein, 10 control) Healthy adults with total cholesterol > 5.2 mmol/L Blinded randomized controlled parallel group trial Mycoprotein-based cookie supplementation (26.9 g dw per day) vs control cookies 8 wk 7.9% ↓ in PC 12.6% ↓ in plasma LDL-C No changes in plasma HDL-C or TGs No changes in Apo A-I or Apo B No differences in BW Turnbull et al (1993)23 13 Healthy females (nonrestrained eaters) Randomized controlled crossover trial Energy-matched mycoprotein-based meal vs chicken-based meal 2 d 24% ↓ in 24-h energy intake on day of meal 16.5% ↓ in 24-h energy intake on day after meal ↓ in prospective food consumption and desire to eat 3 h after meal Burley et al (1993)24 18 Healthy adults Randomized controlled crossover trial Energy-matched mycoprotein-based meal vs chicken-based meal 2 d 18% ↓ in energy intake in subsequent meal ↓ in 24-h energy intake on day of meal (resulting from no compensation after reduction in subsequent meal) No differences in 24-h energy intake on day after meal No overall differences in eating rate or motivation to eat. Significant ↓ in hunger 4 h after meal Nakamura et al (1994)25 15 Healthy males Randomized parallel group trial Mycoprotein-based cookies/crisps supplementation (18 g or 24 g dw per day) 8 wk 4.3% ↓ in PC in 24-g mycoprotein group Ishikawa (1995)26 37 Hypercholesteremic patients, with total cholesterol > 220 mg/dl Double-blind randomized controlled parallel group trial Mycoprotein-based cookie supplementation (12 g or 24 g dw per day) vs control cookies 4 wk ↓ in PC Homma et al (1995)27 52 Healthy males Randomized crossover trial Mycoprotein-based crisps supplementation (18 g or 24 g dw per day) 4 wk 6.7% ↓ in PC in 24-g mycoprotein group Turnbull & Ward (1995)28 19 Healthy adults Double-blind randomized controlled crossover trial Mycoprotein-based milkshake (20 g dw) vs control milkshake 120 min ↓ in glycemia (13% at 60 min) ↓ in insulinemia (19% at 30 min and 36% at 60 min) Williamson et al (2006)29 42 Overweight premenopausal women Randomized controlled crossover trial Mycoprotein-based preload meal vs tofu or chicken-based preload meals before lunch 1 d 12.3% ↓ in energy intake at lunch 20 min after mycoprotein preload when compared with chicken preload No difference in intake at dinner (no compensation) No differences in subjective ratings of hunger or satiety Ruxton & McMillan (2010)30 31 (21 mycoprotein, 10 control) Healthy adults Controlled parallel group trial Mycoprotein-based diet (≥ 88 g ww per day; 21 g dw per day) vs animal-based diet 6 wk ↓ in PC in individuals with baseline cholesterol ≥ 4.19 mmol/L No changes in TC, LDL-C, HDL-C, TGs, glucose, BP, BMI, or WC for sample as a whole Bottin et al (2016)31 Part A: 36 Part B: 14 Overweight and obese adults Single-blind randomized controlled crossover trial Part A: Energy-matched mycoprotein-based preload meal (44 g, 88 g, or 132 g ww) vs chicken-based meal (equivalent amount of chicken and macronutrient matched at each protein content) Part B: Macronutrient-matched mycoprotein-based meal (132 g ww) vs chicken-based meal 180 min 10% ↓ in energy intake at lunch after high mycoprotein preload when compared with high chicken preload 9% ↓ in 24-h energy intake following mycoprotein ingestion 8%, 12%, and 21% ↓ in insulin iAUC after low, medium, and high mycoprotein preload, respectively. 21% and16% ↓ in insulinogenic and disposition indices, respectively, following mycoprotein ingestion 9% ↑ in Matsuda Index following mycoprotein ingestion No differences in appetite ratings No differences in postprandial glucose concentrations No differences in plasma GLP-1 or PYY No differences in gastric emptying No differences in resting energy expenditure and substrate utilization Dunlop et al (2017)15 12 Healthy men Single-blind randomized controlled crossover trial Mycoprotein-based drinks (20 g, 40 g, 60 g, and 80 g dw) vs milk protein drink 240 min Equivalent postprandial AA bioavailability between protein-matched amounts of mycoprotein and milk protein. Slower but more sustained hyperinsulinemia and hyperaminoacidemia compared with milk when protein matched. Dose-response effects on all parameters until 60–80 g of mycoprotein consumed Reference . No. of participants . Participants . Type of study . Type of intervention . Duration of intervention . Study findings . Udall et al (1984)20 100 Healthy adults Double-blind crossover trial Mycoprotein-based cookie supplementation (20 g dw per day) vs control cookies 30 d 6.9% ↓ in PC No changes in BW or other blood markers (glucose, urea, nitrogen, sodium, potassium, calcium, phosphorus, uric acid, creatinine, lactic acid dehydrogenase, alkaline phosphatase, amylase, SGOT, total protein, albumin, TGs, CBC) No changes in urine markers (pH, glucose, protein, ketones, white and red blood cells) Turnbull et al (1990)21 17 (9 mycoprotein, 8 control) Healthy adults with total cholesterol between 5.2 and 6.2 mmol/L Randomized controlled parallel group trial Mycoprotein (≈ 191 g Quorn per day) vs meat during a fully controlled diet 3 wk 13% ↓ in PC 9% ↓ in plasma LDL-C (12% ↑ in control group) 12% ↑ in plasma HDL-C (11% ↓ in control group) 53% ↓ in TGs (in both groups) No differences in BW or BP No changes in fasting insulin and glucose No changes in Apo A-I or Apo B Turnbull et al (1992)22 21 (11 mycoprotein, 10 control) Healthy adults with total cholesterol > 5.2 mmol/L Blinded randomized controlled parallel group trial Mycoprotein-based cookie supplementation (26.9 g dw per day) vs control cookies 8 wk 7.9% ↓ in PC 12.6% ↓ in plasma LDL-C No changes in plasma HDL-C or TGs No changes in Apo A-I or Apo B No differences in BW Turnbull et al (1993)23 13 Healthy females (nonrestrained eaters) Randomized controlled crossover trial Energy-matched mycoprotein-based meal vs chicken-based meal 2 d 24% ↓ in 24-h energy intake on day of meal 16.5% ↓ in 24-h energy intake on day after meal ↓ in prospective food consumption and desire to eat 3 h after meal Burley et al (1993)24 18 Healthy adults Randomized controlled crossover trial Energy-matched mycoprotein-based meal vs chicken-based meal 2 d 18% ↓ in energy intake in subsequent meal ↓ in 24-h energy intake on day of meal (resulting from no compensation after reduction in subsequent meal) No differences in 24-h energy intake on day after meal No overall differences in eating rate or motivation to eat. Significant ↓ in hunger 4 h after meal Nakamura et al (1994)25 15 Healthy males Randomized parallel group trial Mycoprotein-based cookies/crisps supplementation (18 g or 24 g dw per day) 8 wk 4.3% ↓ in PC in 24-g mycoprotein group Ishikawa (1995)26 37 Hypercholesteremic patients, with total cholesterol > 220 mg/dl Double-blind randomized controlled parallel group trial Mycoprotein-based cookie supplementation (12 g or 24 g dw per day) vs control cookies 4 wk ↓ in PC Homma et al (1995)27 52 Healthy males Randomized crossover trial Mycoprotein-based crisps supplementation (18 g or 24 g dw per day) 4 wk 6.7% ↓ in PC in 24-g mycoprotein group Turnbull & Ward (1995)28 19 Healthy adults Double-blind randomized controlled crossover trial Mycoprotein-based milkshake (20 g dw) vs control milkshake 120 min ↓ in glycemia (13% at 60 min) ↓ in insulinemia (19% at 30 min and 36% at 60 min) Williamson et al (2006)29 42 Overweight premenopausal women Randomized controlled crossover trial Mycoprotein-based preload meal vs tofu or chicken-based preload meals before lunch 1 d 12.3% ↓ in energy intake at lunch 20 min after mycoprotein preload when compared with chicken preload No difference in intake at dinner (no compensation) No differences in subjective ratings of hunger or satiety Ruxton & McMillan (2010)30 31 (21 mycoprotein, 10 control) Healthy adults Controlled parallel group trial Mycoprotein-based diet (≥ 88 g ww per day; 21 g dw per day) vs animal-based diet 6 wk ↓ in PC in individuals with baseline cholesterol ≥ 4.19 mmol/L No changes in TC, LDL-C, HDL-C, TGs, glucose, BP, BMI, or WC for sample as a whole Bottin et al (2016)31 Part A: 36 Part B: 14 Overweight and obese adults Single-blind randomized controlled crossover trial Part A: Energy-matched mycoprotein-based preload meal (44 g, 88 g, or 132 g ww) vs chicken-based meal (equivalent amount of chicken and macronutrient matched at each protein content) Part B: Macronutrient-matched mycoprotein-based meal (132 g ww) vs chicken-based meal 180 min 10% ↓ in energy intake at lunch after high mycoprotein preload when compared with high chicken preload 9% ↓ in 24-h energy intake following mycoprotein ingestion 8%, 12%, and 21% ↓ in insulin iAUC after low, medium, and high mycoprotein preload, respectively. 21% and16% ↓ in insulinogenic and disposition indices, respectively, following mycoprotein ingestion 9% ↑ in Matsuda Index following mycoprotein ingestion No differences in appetite ratings No differences in postprandial glucose concentrations No differences in plasma GLP-1 or PYY No differences in gastric emptying No differences in resting energy expenditure and substrate utilization Dunlop et al (2017)15 12 Healthy men Single-blind randomized controlled crossover trial Mycoprotein-based drinks (20 g, 40 g, 60 g, and 80 g dw) vs milk protein drink 240 min Equivalent postprandial AA bioavailability between protein-matched amounts of mycoprotein and milk protein. Slower but more sustained hyperinsulinemia and hyperaminoacidemia compared with milk when protein matched. Dose-response effects on all parameters until 60–80 g of mycoprotein consumed Abbreviations and symbols: AA, amino acid; APO, apolipoprotein; BMI, body mass index; BP, blood pressure; CBC, complete blood count; dw, dry weight; GLP-1, glucagon-like peptide 1; HDL-C, high-density lipoprotein cholesterol; iAUC, incremental area under the curve; LDL-C, low-density lipoprotein cholesterol; PC, plasma cholesterol; PYY, peptide YY/peptide tyrosine tyrosine; SGOT, serum glutamic oxaloacetic transaminase; TC, total cholesterol; TGs, triglycerides; WC, waist circumference; ww, wet weight; ↑, increase; ↓, decrease. Open in new tab Table 3 Human studies investigating the metabolic effects of mycoprotein Reference . No. of participants . Participants . Type of study . Type of intervention . Duration of intervention . Study findings . Udall et al (1984)20 100 Healthy adults Double-blind crossover trial Mycoprotein-based cookie supplementation (20 g dw per day) vs control cookies 30 d 6.9% ↓ in PC No changes in BW or other blood markers (glucose, urea, nitrogen, sodium, potassium, calcium, phosphorus, uric acid, creatinine, lactic acid dehydrogenase, alkaline phosphatase, amylase, SGOT, total protein, albumin, TGs, CBC) No changes in urine markers (pH, glucose, protein, ketones, white and red blood cells) Turnbull et al (1990)21 17 (9 mycoprotein, 8 control) Healthy adults with total cholesterol between 5.2 and 6.2 mmol/L Randomized controlled parallel group trial Mycoprotein (≈ 191 g Quorn per day) vs meat during a fully controlled diet 3 wk 13% ↓ in PC 9% ↓ in plasma LDL-C (12% ↑ in control group) 12% ↑ in plasma HDL-C (11% ↓ in control group) 53% ↓ in TGs (in both groups) No differences in BW or BP No changes in fasting insulin and glucose No changes in Apo A-I or Apo B Turnbull et al (1992)22 21 (11 mycoprotein, 10 control) Healthy adults with total cholesterol > 5.2 mmol/L Blinded randomized controlled parallel group trial Mycoprotein-based cookie supplementation (26.9 g dw per day) vs control cookies 8 wk 7.9% ↓ in PC 12.6% ↓ in plasma LDL-C No changes in plasma HDL-C or TGs No changes in Apo A-I or Apo B No differences in BW Turnbull et al (1993)23 13 Healthy females (nonrestrained eaters) Randomized controlled crossover trial Energy-matched mycoprotein-based meal vs chicken-based meal 2 d 24% ↓ in 24-h energy intake on day of meal 16.5% ↓ in 24-h energy intake on day after meal ↓ in prospective food consumption and desire to eat 3 h after meal Burley et al (1993)24 18 Healthy adults Randomized controlled crossover trial Energy-matched mycoprotein-based meal vs chicken-based meal 2 d 18% ↓ in energy intake in subsequent meal ↓ in 24-h energy intake on day of meal (resulting from no compensation after reduction in subsequent meal) No differences in 24-h energy intake on day after meal No overall differences in eating rate or motivation to eat. Significant ↓ in hunger 4 h after meal Nakamura et al (1994)25 15 Healthy males Randomized parallel group trial Mycoprotein-based cookies/crisps supplementation (18 g or 24 g dw per day) 8 wk 4.3% ↓ in PC in 24-g mycoprotein group Ishikawa (1995)26 37 Hypercholesteremic patients, with total cholesterol > 220 mg/dl Double-blind randomized controlled parallel group trial Mycoprotein-based cookie supplementation (12 g or 24 g dw per day) vs control cookies 4 wk ↓ in PC Homma et al (1995)27 52 Healthy males Randomized crossover trial Mycoprotein-based crisps supplementation (18 g or 24 g dw per day) 4 wk 6.7% ↓ in PC in 24-g mycoprotein group Turnbull & Ward (1995)28 19 Healthy adults Double-blind randomized controlled crossover trial Mycoprotein-based milkshake (20 g dw) vs control milkshake 120 min ↓ in glycemia (13% at 60 min) ↓ in insulinemia (19% at 30 min and 36% at 60 min) Williamson et al (2006)29 42 Overweight premenopausal women Randomized controlled crossover trial Mycoprotein-based preload meal vs tofu or chicken-based preload meals before lunch 1 d 12.3% ↓ in energy intake at lunch 20 min after mycoprotein preload when compared with chicken preload No difference in intake at dinner (no compensation) No differences in subjective ratings of hunger or satiety Ruxton & McMillan (2010)30 31 (21 mycoprotein, 10 control) Healthy adults Controlled parallel group trial Mycoprotein-based diet (≥ 88 g ww per day; 21 g dw per day) vs animal-based diet 6 wk ↓ in PC in individuals with baseline cholesterol ≥ 4.19 mmol/L No changes in TC, LDL-C, HDL-C, TGs, glucose, BP, BMI, or WC for sample as a whole Bottin et al (2016)31 Part A: 36 Part B: 14 Overweight and obese adults Single-blind randomized controlled crossover trial Part A: Energy-matched mycoprotein-based preload meal (44 g, 88 g, or 132 g ww) vs chicken-based meal (equivalent amount of chicken and macronutrient matched at each protein content) Part B: Macronutrient-matched mycoprotein-based meal (132 g ww) vs chicken-based meal 180 min 10% ↓ in energy intake at lunch after high mycoprotein preload when compared with high chicken preload 9% ↓ in 24-h energy intake following mycoprotein ingestion 8%, 12%, and 21% ↓ in insulin iAUC after low, medium, and high mycoprotein preload, respectively. 21% and16% ↓ in insulinogenic and disposition indices, respectively, following mycoprotein ingestion 9% ↑ in Matsuda Index following mycoprotein ingestion No differences in appetite ratings No differences in postprandial glucose concentrations No differences in plasma GLP-1 or PYY No differences in gastric emptying No differences in resting energy expenditure and substrate utilization Dunlop et al (2017)15 12 Healthy men Single-blind randomized controlled crossover trial Mycoprotein-based drinks (20 g, 40 g, 60 g, and 80 g dw) vs milk protein drink 240 min Equivalent postprandial AA bioavailability between protein-matched amounts of mycoprotein and milk protein. Slower but more sustained hyperinsulinemia and hyperaminoacidemia compared with milk when protein matched. Dose-response effects on all parameters until 60–80 g of mycoprotein consumed Reference . No. of participants . Participants . Type of study . Type of intervention . Duration of intervention . Study findings . Udall et al (1984)20 100 Healthy adults Double-blind crossover trial Mycoprotein-based cookie supplementation (20 g dw per day) vs control cookies 30 d 6.9% ↓ in PC No changes in BW or other blood markers (glucose, urea, nitrogen, sodium, potassium, calcium, phosphorus, uric acid, creatinine, lactic acid dehydrogenase, alkaline phosphatase, amylase, SGOT, total protein, albumin, TGs, CBC) No changes in urine markers (pH, glucose, protein, ketones, white and red blood cells) Turnbull et al (1990)21 17 (9 mycoprotein, 8 control) Healthy adults with total cholesterol between 5.2 and 6.2 mmol/L Randomized controlled parallel group trial Mycoprotein (≈ 191 g Quorn per day) vs meat during a fully controlled diet 3 wk 13% ↓ in PC 9% ↓ in plasma LDL-C (12% ↑ in control group) 12% ↑ in plasma HDL-C (11% ↓ in control group) 53% ↓ in TGs (in both groups) No differences in BW or BP No changes in fasting insulin and glucose No changes in Apo A-I or Apo B Turnbull et al (1992)22 21 (11 mycoprotein, 10 control) Healthy adults with total cholesterol > 5.2 mmol/L Blinded randomized controlled parallel group trial Mycoprotein-based cookie supplementation (26.9 g dw per day) vs control cookies 8 wk 7.9% ↓ in PC 12.6% ↓ in plasma LDL-C No changes in plasma HDL-C or TGs No changes in Apo A-I or Apo B No differences in BW Turnbull et al (1993)23 13 Healthy females (nonrestrained eaters) Randomized controlled crossover trial Energy-matched mycoprotein-based meal vs chicken-based meal 2 d 24% ↓ in 24-h energy intake on day of meal 16.5% ↓ in 24-h energy intake on day after meal ↓ in prospective food consumption and desire to eat 3 h after meal Burley et al (1993)24 18 Healthy adults Randomized controlled crossover trial Energy-matched mycoprotein-based meal vs chicken-based meal 2 d 18% ↓ in energy intake in subsequent meal ↓ in 24-h energy intake on day of meal (resulting from no compensation after reduction in subsequent meal) No differences in 24-h energy intake on day after meal No overall differences in eating rate or motivation to eat. Significant ↓ in hunger 4 h after meal Nakamura et al (1994)25 15 Healthy males Randomized parallel group trial Mycoprotein-based cookies/crisps supplementation (18 g or 24 g dw per day) 8 wk 4.3% ↓ in PC in 24-g mycoprotein group Ishikawa (1995)26 37 Hypercholesteremic patients, with total cholesterol > 220 mg/dl Double-blind randomized controlled parallel group trial Mycoprotein-based cookie supplementation (12 g or 24 g dw per day) vs control cookies 4 wk ↓ in PC Homma et al (1995)27 52 Healthy males Randomized crossover trial Mycoprotein-based crisps supplementation (18 g or 24 g dw per day) 4 wk 6.7% ↓ in PC in 24-g mycoprotein group Turnbull & Ward (1995)28 19 Healthy adults Double-blind randomized controlled crossover trial Mycoprotein-based milkshake (20 g dw) vs control milkshake 120 min ↓ in glycemia (13% at 60 min) ↓ in insulinemia (19% at 30 min and 36% at 60 min) Williamson et al (2006)29 42 Overweight premenopausal women Randomized controlled crossover trial Mycoprotein-based preload meal vs tofu or chicken-based preload meals before lunch 1 d 12.3% ↓ in energy intake at lunch 20 min after mycoprotein preload when compared with chicken preload No difference in intake at dinner (no compensation) No differences in subjective ratings of hunger or satiety Ruxton & McMillan (2010)30 31 (21 mycoprotein, 10 control) Healthy adults Controlled parallel group trial Mycoprotein-based diet (≥ 88 g ww per day; 21 g dw per day) vs animal-based diet 6 wk ↓ in PC in individuals with baseline cholesterol ≥ 4.19 mmol/L No changes in TC, LDL-C, HDL-C, TGs, glucose, BP, BMI, or WC for sample as a whole Bottin et al (2016)31 Part A: 36 Part B: 14 Overweight and obese adults Single-blind randomized controlled crossover trial Part A: Energy-matched mycoprotein-based preload meal (44 g, 88 g, or 132 g ww) vs chicken-based meal (equivalent amount of chicken and macronutrient matched at each protein content) Part B: Macronutrient-matched mycoprotein-based meal (132 g ww) vs chicken-based meal 180 min 10% ↓ in energy intake at lunch after high mycoprotein preload when compared with high chicken preload 9% ↓ in 24-h energy intake following mycoprotein ingestion 8%, 12%, and 21% ↓ in insulin iAUC after low, medium, and high mycoprotein preload, respectively. 21% and16% ↓ in insulinogenic and disposition indices, respectively, following mycoprotein ingestion 9% ↑ in Matsuda Index following mycoprotein ingestion No differences in appetite ratings No differences in postprandial glucose concentrations No differences in plasma GLP-1 or PYY No differences in gastric emptying No differences in resting energy expenditure and substrate utilization Dunlop et al (2017)15 12 Healthy men Single-blind randomized controlled crossover trial Mycoprotein-based drinks (20 g, 40 g, 60 g, and 80 g dw) vs milk protein drink 240 min Equivalent postprandial AA bioavailability between protein-matched amounts of mycoprotein and milk protein. Slower but more sustained hyperinsulinemia and hyperaminoacidemia compared with milk when protein matched. Dose-response effects on all parameters until 60–80 g of mycoprotein consumed Abbreviations and symbols: AA, amino acid; APO, apolipoprotein; BMI, body mass index; BP, blood pressure; CBC, complete blood count; dw, dry weight; GLP-1, glucagon-like peptide 1; HDL-C, high-density lipoprotein cholesterol; iAUC, incremental area under the curve; LDL-C, low-density lipoprotein cholesterol; PC, plasma cholesterol; PYY, peptide YY/peptide tyrosine tyrosine; SGOT, serum glutamic oxaloacetic transaminase; TC, total cholesterol; TGs, triglycerides; WC, waist circumference; ww, wet weight; ↑, increase; ↓, decrease. Open in new tab MYCOPROTEIN, DIETARY FIBER, AND CARDIOMETABOLIC HEALTH During an initial human investigation aimed at establishing the tolerability of mycoprotein, an interesting ancillary observation was made.20 Human volunteers who consumed 20 g of mycoprotein (dry weight) per day for 30 days (consumed as supplemental cookies) showed an approximately 7% decrease in blood cholesterol concentrations (from 4.86 to 4.53 mmol/L),20 which replicated earlier findings in animals.32–34 Early research was then focused on the potential health effects of the dietary fiber in mycoprotein. This stemmed from a body of epidemiological studies showing that higher fiber intakes (typically from fruit, vegetables, and cereals) are associated with reduced blood cholesterol concentrations, improved blood lipid profiles, and reduced incidence of myocardial infarction and coronary heart disease.35–38 Such findings have been confirmed by intervention studies in which increased consumption of dietary fiber has been reported to improve peripheral insulin sensitivity and to lower blood cholesterol concentrations and glycated hemoglobin (HBA1c) in both healthy individuals and patients with type 2 diabetes.39,40 Follow-up studies focusing on mycoprotein consumption and cardiometabolic health confirmed and extended these findings with regard to blood lipid profiles.21,22 Turnbull et al21 performed a 3-week dietary intervention study in which participants consumed 191 g of mycoprotein-containing products (≈ 40 g of mycoprotein, dry weight) per day as part of a fully controlled and laboratory-supervised diet aimed at maintaining energy balance in individuals with mildly elevated blood cholesterol concentrations. The mycoprotein intervention resulted in reduced total cholesterol (from 5.54 to 4.81 mmol/L; 13% decrease) and low-density lipoprotein cholesterol (LDL-C) (from 4.16 to 3.78 mmol/L; 9% decrease) and increased high-density lipoprotein cholesterol (HDL-C) (from 0.58 to 0.65 mmol/L; 12% increase). These results were especially striking, considering the control group generally showed opposite responses (as opposed to no change). Given that the lipid composition and the energy, macronutrient, and cholesterol contents of the diets were similar across groups, it was assumed that the fiber content was responsible for the changes in blood lipids. The increased dietary fiber content could have conceivably exerted its cholesterol-lowering effect by altering LDL-C synthesis/degradation or cholesterol clearance in peripheral tissues or by increasing the binding of fiber to neutral sterols, cholesterol, or bile acids in the intestine, resulting in a decreased amount of cholesterol entering the circulating pool. However, it is noteworthy that the beneficial effects of higher-fiber diets on circulating cholesterol concentrations do not always extend to improvements in the specific lipid subfractions of LDL-C and HDL-C.41 It is thus interesting to ponder whether the type, rather than simply the amount, of dietary fiber contained within mycoprotein may, at least in part, explain the beneficial effects of mycoprotein consumption on circulating cholesterol concentrations. Dietary fiber in mycoprotein is composed of two-thirds β-glucan and one-third chitin, which together form a fibrous insoluble matrix that is relatively rare in more traditional food sources. In their follow-up work focused on fiber type, Turnbull et al22 reported the effects of mycoprotein consumption in free-living individuals (a 0.95 mmol/L, or a 16% reduction, and a 0.34 mmol/L, or 21% reduction, in total cholesterol and LDL-C, respectively) to be similar to those observed previously, even though overall energy, macronutrients, and fiber (around 6 g) intake remained the same across groups. Recent in vitro investigations have dug deeper mechanistically and are beginning to shed light on potential mechanisms by which the specific fiber profile of mycoprotein may affect the gut microbiota to bring about these cholesterol-lowering effects in humans. Upon entering the large intestine, protein and dietary fibers become available for fermentation by the gut microbiota.42 Fermentation of dietary fibers leads to the production of short-chain fatty acids (SCFAs), primarily acetate, propionate, and butyrate, in a molar ratio of approximately 60:20:20.43 Fermentation of protein-derived amino acids leads to the production of phenols, amines, ammonia, branched-chain fatty acids, and SCFAs. The gut microbiota prioritizes dietary fiber fermentation over protein fermentation, and when fiber is undergoing active fermentation, the fate of dietary protein-derived amino acids is bacterial cell biomass, as opposed to metabolism. The switch from fiber fermentation to protein fermentation has also been shown to have profound effects on the composition of the gut microbiota.44 Production of SCFAs, and of propionate in particular, has been shown to reduce hepatic cholesterol synthesis via inhibition of β-hydroxy-β-methylglutaryl coenzyme A reductase (the rate-limiting enzyme in cholesterol synthesis)45 and suppress adipose tissue lipolysis.46 In human studies using inulin propionate ester, which delivers propionate directly to the large intestine, propionate has been demonstrated to reduce LDL-C and improve liver function and insulin sensitivity.47,48 However, current evidence shows the effect of propionate to be inconsistent, with another study suggesting that propionate consumption leads to insulin resistance and compensatory hyperinsulinemia.49 In vitro colonic models have shown that mycoprotein and its purified dietary fiber are fermentable, capable of producing SCFAs.50 Both mycoprotein and purified mycoprotein dietary fiber produce increased propionate and butyrate at the expense of acetate, and increasing colonic propionate production inhibits the incorporation of plasma acetate into cholesterol.51 Consequently, available data now suggest that the digestive and metabolic properties of the unique fiber profile present in mycoprotein warrant future (in vivo) research. The beneficial metabolic effects of mycoprotein consumption have also been shown to extend to acute postprandial glycemic control.28,31 It was reported that 20 g of mycoprotein (dry weight) consumed during an oral glucose tolerance test resulted in reduced postprandial glycemia and insulinemia compared with an isonitrogenous, isoenergetic control condition (soy and skimmed milk) in healthy, young adults.28 A recent study in overweight adults reported that mycoprotein consumption (≈ 40 g, dry weight), compared with an energy- and macronutrient-matched chicken meal, reduced postprandial insulinemia, but not glycemia.31 Again, the causative mechanism is likely linked to the amount (4 g and 7 g, respectively) and type of fiber contained in mycoprotein in these 2 studies, as viscous polysaccharides can reduce postprandial glycemia and insulinemia,52 and 5 g of β-glucan has previously been shown to modulate glycemia and insulinemia when consumed with a high carbohydrate load.53 Though the chitin-glucan matrix is insoluble and not viscous, chitin is likely to undergo alkaline deacetylation to produce the viscous polysaccharide chitosan at some point in the gastrointestinal tract. In turn, this may confer resistance to the flow induced by gastrointestinal motility, reducing contact time in the small intestine and resulting in slower gastric emptying and, consequently, nutrient absorption.54 Irrespective of the mechanism, no data are yet available to confirm whether these acute effects on postprandial glycemia extend to robust changes in insulin sensitivity or habitual glycemic control when mycoprotein is incorporated into the daily diet. MYCOPROTEIN AND WEIGHT MANAGEMENT The growing obesity epidemic, along with its associated health complications,55 requires nutritional approaches to induce and sustain weight loss. Although weight loss via caloric restriction under laboratory conditions is relatively straightforward to achieve,56 it tends to be more difficult under free-living conditions. Furthermore, subsequent weight regain appears to be the major barrier to longer-term weight management.57 The primary reasons for these difficulties include a lack of satiety while maintaining an energy deficit58 and a decline in the basal metabolic rate as a result of loss of muscle mass.59,60 Diets relatively high in protein (generally those diets in which absolute protein intake is simply maintained while an energy deficit is created by restricting carbohydrates or fats) have been suggested as a potential solution to these issues.8,57 For instance, when volunteers are subjected to ad libitum weight loss diets (ie, more representative of free-living attempts at weight loss), those consuming diets higher in protein generally lose body mass and maintain this loss more effectively than those on lower-protein diets.61,62 This seems primarily attributable to the satiating effects of protein ingestion, which results in overall lower energy intake,61 since isoenergetically controlled weight loss interventions show equivalent weight loss irrespective of protein content.61,62 It is also true that higher-protein diets increase overall daily energy expenditure as a result of enhanced diet-induced thermogenesis and energy expenditure during sleep, effects that occur regardless of the type of protein consumed.63,64 Furthermore, during isoenergetically controlled weight loss studies, it has typically been shown that higher-protein diets increase the ratio of fat mass loss to lean mass loss, which constitutes overall body weight loss.65 The effects of dietary protein intake during weight loss on satiety, daily energy expenditure, and lean mass retention, taken together, likely explain the role of dietary protein in effective long-term weight loss and management.8,57,61 Mycoprotein consumption has been shown to induce an acute thermogenic response, similar to that seen following the ingestion of other (animal) protein sources,15 and therefore would presumably contribute to overall daily energy expenditure during a weight loss regimen, as described above. Additionally, mycoprotein and most mycoprotein-containing products have a low energy density. The consumption of foods with a low energy density is positively associated with reduced ad libitum energy intake and positive weight management outcomes.66 The substitution of high-energy-density foods for mycoprotein-containing products may be an effective tool to manipulate the energy density of a meal or diet. As a low-energy-density, high-protein food source, mycoprotein would also have theoretical value in a diet aimed at maintaining protein intake in an effort to retain lean tissue during an energy deficit. The effects of mycoprotein on satiety are also of particular interest. It has been shown previously that protein sources differ in their capacity to affect satiety.64 For example, gelatin protein provided as a single meal67 or as a primary protein source over a 36-hour experimental period64 was reported to suppress appetite to a greater extent when compared with isonitrogenous milk protein equivalents, which the authors suggested may be related to the effects of amino acid composition in the central nervous system. Differences in sensory characteristics, such as greater viscosity and creaminess, may also play a role in increasing satiety and reducing energy intake.68–71 Interestingly, Turnbull et al23 demonstrated that consumption of a mycoprotein meal resulted in acute appetite suppression and a subsequent reduction in ad libitum food consumption for the remainder of the day (by 24%) and for the following day (by 17%) when compared with an isoenergetic and isonitrogenous chicken meal. Similar findings were reproduced by Burley et al24 and Williamson et al29 when study participants consumed approximately 30 g and 10 g (dry weight) of mycoprotein, respectively. Equivalent satiety between mycoprotein and milk protein has also been reported.15 In the Turnbull et al23 study, these effects were attributed to the greater dietary fiber content of the mycoprotein meal (since the meals were equivalent for energy and protein intake, fiber was necessarily higher) compared with the chicken meal. Additionally, given the relatively small difference in fiber content between conditions (10 g vs 17 g), the authors also suggested these effects may be attributable to either the specific type of fiber, which may be particularly potent, or slower gastric emptying. Interestingly, both aspects could ultimately act by modulating postprandial (neuro)endocrine responses. However, a recent report of similar increased satiety effects of mycoprotein compared with chicken in overweight and obese individuals does not support a role of postprandial secretion of the gut peptide YY (PYY) or the hormone glucagon-like peptide 1 (GLP-1) (both commonly purported to play a role in appetite suppression following food intake) as a causative mechanism.31 It is possible that various metabolites associated with the partial fermentation of dietary fibers may explain the potent appetite-suppressing effect of mycoprotein.31 For example, the SCFA propionate has been shown to induce secretion of PYY and GLP-1 in humans in studies of acute ingestion and may, in part, explain the short-term appetite-regulating effects of some dietary fibers.47 Both mycoprotein and mycoprotein-derived dietary fiber promote the production of propionate, but the relevance of this mechanism in explaining the effects on appetite regulation remains to be fully elucidated.50 Irrespective of the mechanism, the effects of mycoprotein on satiety and thermogenesis, along with the high-protein/low-energy content of mycoprotein, position this food source as an intriguing approach to support a diet (under ad libitum conditions) aimed at weight loss or maintenance. Also worthy of note, diets with a lower glycemic index have independently been shown to improve weight maintenance following weight loss during energy restriction.57 The capacity of mycoprotein to lower the glycemic load of a meal or habitual diet adds an additional line of inquiry about its potential utility in weight management. Well-controlled, longer-term laboratory weight loss studies comparing mycoprotein with other protein sources are warranted. MYCOPROTEIN AND SKELETAL MUSCLE ADAPTATION Adequate dietary protein intake is required for skeletal muscle mass maintenance and reconditioning. Skeletal muscle mass and the quality of skeletal protein are maintained (or improved) through dynamic fluctuations in the rates of muscle protein synthesis and breakdown. In the overnight, fasted state, rates of muscle protein breakdown exceed rates of muscle protein synthesis, leading to net loss of muscle protein.72 Protein ingestion transiently (for 2–5 h) increases the rates of muscle protein synthesis,73 primarily because of elevated plasma essential amino acids,74 of which leucine is of particular relevance.10,75 Protein ingestion also stimulates the secretion of pancreatic insulin, which inhibits muscle protein breakdown,76 thus contributing to net accretion of muscle protein (the so-called anabolic response) in the postprandial state and offsetting protein losses from the fasting state. It is these diurnal oscillations in muscle protein balance that ultimately allow individuals to maintain muscle mass. Individuals performing structured and prolonged physical activity will elicit adaptive responses within skeletal muscle, such as increases in muscle mass, muscle quality, contractile function, and muscle oxidative capacity. Performing physical activity stimulates the rate of muscle protein synthesis and, to a lesser extent, the rate of muscle protein breakdown, improving muscle protein balance for up to 48 hours.77 The accumulation of periods of exercise-induced muscle protein accretion ultimately drives skeletal muscle reconditioning. Following resistance training, this response primarily comprises the synthesis of myofibrillar proteins to support strength- and mass-related adaptations.78 Conversely, in response to endurance exercise, it is predominantly mitochondrial proteins that are synthesized to facilitate improved oxidative capacity.78 Consuming dietary protein in close temporal proximity to physical activity is an established strategy to further augment the muscle protein synthetic response compared with either stimulus alone.79,80 As a result, strategically (and modestly) increasing dietary protein consumption during prolonged training augments the skeletal muscle adaptive response to exercise training.81,82 Since rates of postprandial muscle protein breakdown appear to be maximally inhibited with only mild elevations in circulating insulin,83 the anabolic potential of (postexercise) dietary protein ingestion is assumed to be contingent on the protein’s capacity to stimulate muscle protein synthesis. Animal-derived proteins typically have high bioavailability, and rapid and sustained postprandial aminoacidemia and leucinemia are observed following their ingestion.10,84–86 As a result, animal-derived dietary protein sources have been shown to be superior to plant-based protein sources in their capacity to stimulate muscle protein synthesis rates in humans.86,87 However, to date, wheat and soy (both relatively low in leucine and essential amino acids88) are the only non–animal-derived protein sources to be evaluated for their anabolic potential. Mycoprotein is rich in essential amino acids (Table 2) (≈ 41% of total protein) and relatively high in leucine (≈ 6% of total protein) and possesses a high PDCAAS (protein digestibility–corrected amino acid score, an indirect indication of a protein’s digestibility) of 0.99. The in vivo amino acid bioavailability of mycoprotein was recently compared with that of isolated milk protein.15 Milk protein was selected as the control comparator because it contains high proportions of essential amino acids (≈ 49% of total protein) and leucine (≈ 11% of total protein), has a PDCAAS of 1.0, and, consequently, is typically thought of as a nearly gold standard protein source with respect to its potency to stimulate muscle protein synthesis89 and optimize training adaptations.90 The findings showed that, in healthy young men, the bioavailability of essential amino acids and leucine in the hours following ingestion of protein-matched boluses of milk protein was equivalent to that of mycoprotein (though less rapid and more sustained with mycoprotein ingestion).15 It is of note that, to match these conditions for protein content, approximately double the mass (and energy) of mycoprotein was consumed, since mycoprotein is a whole food source. It was also found that the amino acid bioavailability of mycoprotein increases in a dose-response fashion until between 60 g and 80 g of mycoprotein (ie, 27–36 g of protein; 2.1–2.9 g of leucine) is consumed. Therefore, it seems likely that mycoprotein would stimulate a robust muscle protein synthetic response and, when ingested in larger quantities, would be an alternative protein source to support muscle tissue reconditioning during prolonged training. The magnitude of this response, however, when compared with that induced by other protein sources, would presumably depend on whether the overall systemic availability of (essential) amino acids or the speed at which amino acids become available is the more crucial regulatory factor.84,91 An interesting additional consideration is that mycoprotein represents a whole food source, rather than an isolated protein. The latter has generally been employed in studies addressing postprandial responses to muscle protein synthesis. While co-ingestion of carbohydrates or fats with isolated protein does not seem to modulate the postprandial response to muscle protein synthesis,92–94 emerging data indicate that protein consumed within a whole food source may confer an anabolic advantage.95,96 It is not clear whether such effects are attributable to differences in energy or macro-/micronutrient content or to aspects related to a protein source’s specific food matrix. However, the relevance of evaluating the anabolic response to whole food sources is emerging as a key research area necessary to translate laboratory findings into information to refine dietary protein recommendations.96–98 MYCOPROTEIN AND SARCOPENIA In concert with a rising population worldwide, global demographics indicate the number of individuals aged 60 years and older will triple by the year 2050, with the fastest-growing subpopulation being those over 85 years.99 A key hallmark of aging is the progressive loss of skeletal muscle mass and strength (termed sarcopenia), accompanied by reduced aerobic capacity.100 The association between muscle loss (mass and quality) and increased incidence of falls, fractures, metabolic disease, and other health complications indicates that the burden of our aging society on healthcare systems will increase dramatically over the coming decades. Importantly, it also underlines the critical role that skeletal muscle mass and quality play in healthy aging. Since rates of basal, fasted muscle protein synthesis and breakdown do not appear to differ between healthy young and older adults,11,101–104 and in an effort to explain the physiological mechanisms responsible for age-related sarcopenia, research has recently focused on the anabolic response to food intake. Numerous studies have demonstrated a blunted muscle protein synthetic response to protein ingestion in older adults,11,102,105 and this so-called anabolic resistance is now believed to be a key factor underlying age-related sarcopenia. Anabolic resistance can be effectively compensated for on a per-meal basis by consuming protein in close temporal proximity to physical activity,106 by increasing the amount of protein consumed,85,107 or by optimizing the protein source.10,84 On the basis of this mechanistic understanding of senescent muscle protein metabolism, calls from the scientific community to increase the recommended daily amount of protein (and to address optimal types of protein) to support healthy aging are gaining momentum.3,4,108 Moreover, these recommendations are in line with epidemiological studies demonstrating that older adults who consume protein in excess (ie, ≈ 1.2 g per kilogram of body mass) of the Recommended Dietary Allowance (ie, 0.8 g per kilogram of body mass) experience lower rates of decline in muscle mass, strength, and functional capacity.109,110 A pressing question is therefore arising: “From where should this dietary protein to support healthy aging come?” It will become increasingly important to view this question through the potentially competing interests of robust nutritional physiological investigation and the many issues that comprise environmental and government policy. The muscle protein synthetic response of senescent muscle to alternative, non–animal-derived protein sources has scarcely been studied. Whether mycoprotein can provide an effective and sustainable source of dietary protein to support healthy (and active) aging remains to be investigated. While the potential usefulness of mycoprotein is promising, the development of age-related anabolic resistance presents a challenge. It would be expected that a relatively large dose of mycoprotein would be required to maximally stimulate the muscle protein synthetic response in older adults.107 Given that older adults generally display a reduced appetite compared with younger adults, and paired with the potent satiating effect of mycoprotein, it would follow that consuming a sufficient amount of mycoprotein per meal (or over repeated meals to obtain daily intakes) may be challenging. Careful consideration to the other macronutrients that compose a higher-(myco)protein meal would therefore be required. Clearly, future research is warranted to establish whether mycoprotein could be used to support optimal rates of muscle protein synthesis while avoiding a positive or negative energy balance in older adults, thereby having potential as part of a viable strategy to support healthy aging. CONCLUSION Developing sustainable dietary protein sources is a pressing socioeconomic and environmental concern, and there is an obvious need to develop a robust evidence base to inform the use of such alternative sources. Evidence shows that the incorporation of modest amounts of mycoprotein into the diet positively influences certain circulating lipid subfractions, and acute mycoprotein ingestion attenuates postprandial glycemia and insulinemia. These data are striking, as they occur under energy-balanced conditions (ie, they are not an artifact of lower overall energy intake or consequent weight loss). These responses may be mediated by the unique digestive and metabolic properties of the chitin and β-glucan fibers present in mycoprotein, though a comprehensive and in vivo mechanistic understanding is still lacking. It is not known how rapidly circulating cholesterol is affected when mycoprotein is incorporated into the diet, and a full characterization of the lipid subfraction responses is not yet available. Furthermore, whether alterations of acute postprandial glycemic control translate into improved insulin sensitivity or habitual glycemic control when mycoprotein is incorporated in the daily diet is also unclear. Mycoprotein is nutrient-rich and can effectively induce satiety, as evidenced by reduced ad libitum energy intake in several studies, suggesting it may be a useful tool in weight management. It also has potential as a high-quality protein source and as a modulator of postprandial glycemia. In light of these findings, research into the ability of mycoprotein to modulate habitual glycemic control, caloric intake, and weight management is clearly warranted. Emerging data have shown that mycoprotein is a bioavailable and insulinotropic source of protein and would, therefore, be expected to effectively stimulate muscle protein anabolism. Consequently, mycoprotein as a source of dietary protein to stimulate muscle protein synthesis and to promote muscle adaptation or maintenance in various populations (eg, athletes, older adults) is a promising area of future research. Acknowledgments Author contributions. M.O.C.C. and A.J.M. prepared the figures and tables and performed the literature searches. M.O.C.C., A.J.M. and B.T.W. drafted the manuscript. M.V.D., H.C.H., D.J.M., and F.B.S. read and provided intellectual input into draft versions of the manuscript. All authors read and approved the final manuscript. Funding/support B.T.W. has received research grants from Marlow Foods, Stokesley, UK, relating to the content of this review. M.O.C.C. and A.J.M. receive PhD studentship funding from Marlow Foods and the College of Life and Environmental Sciences, University of Exeter. Declaration of interest. B.T.W. is currently receiving grant funding from Marlow Foods, Stokesley, UK, to perform research related to the content of this review. M.O.C.C. and A.J.M. have received PhD funding, in part, from Marlow Foods. M.V.D. is employed on a grant provided by Marlow Foods. The remaining authors have no relevant interests to declare. References 1 Ranganathan J Vennard D Waite R , et al. . Shifting Diets for a Sustainable Food Future: Creating a Sustainable Food Future, Installment Eleven . Washington, DC : World Resources Institute ; 2016 . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC 2 Phillips SM Chevalier S Leidy HJ. Protein “requirements” beyond the RDA: implications for optimizing health . Appl Physiol Nutr Metab. 2016 ; 41 : 565 – 572 . Google Scholar Crossref Search ADS PubMed WorldCat 3 Phillips SM. Nutrition in the elderly: a recommendation for more (evenly distributed) protein? Am J Clin Nutr. 2017 ; 106 : 12 – 13 . Google Scholar Crossref Search ADS PubMed WorldCat 4 Paddon-Jones D Campbell WW Jacques PF , et al. . Protein and healthy aging . Am J Clin Nutr . 2015 ; 101 : 1339S – 1345S . doi: 10.3945/ajcn.114.084061 Google Scholar Crossref Search ADS PubMed WorldCat 5 Wall BT Cermak NM van Loon LJ. Dietary protein considerations to support active aging . Sports Med. 2014 ; 44(suppl 2) : 185–194. OpenURL Placeholder Text WorldCat 6 Phillips SM. Dietary protein requirements and adaptive advantages in athletes . Br J Nutr. 2012 ; 108(suppl 2) : S158 – S167 . Google Scholar Crossref Search ADS PubMed WorldCat 7 Wall BT Morton JP van Loon LJ. Strategies to maintain skeletal muscle mass in the injured athlete: nutritional considerations and exercise mimetics . Eur J Sport Sci . 2015 ; 15 : 53 – 62 . Google Scholar Crossref Search ADS PubMed WorldCat 8 Westerterp-Plantenga MS Lemmens SG Westerterp KR. Dietary protein—its role in satiety, energetics, weight loss and health . Br J Nutr. 2012 ; 108(suppl 2) : S105–S112. OpenURL Placeholder Text WorldCat 9 Burd NA Gorissen SH van Vliet S , et al. . Differences in postprandial protein handling after beef compared with milk ingestion during postexercise recovery: a randomized controlled trial . Am J Clin Nutr . 2015 ; 102 : 828 – 836 . Google Scholar Crossref Search ADS PubMed WorldCat 10 Wall BT Hamer HM de Lange A , et al. . Leucine co-ingestion improves post-prandial muscle protein accretion in elderly men . Clin Nutr . 2013 ; 32 : 412 – 419 . Google Scholar Crossref Search ADS PubMed WorldCat 11 Wall BT Gorissen SH Pennings B , et al. . Aging is accompanied by a blunted muscle protein synthetic response to protein ingestion . PLoS One. 2015 ; 10 : E 0140903. Google Scholar Crossref Search ADS WorldCat 12 Nadathur SR Wanasundara JPD Scanlin L, eds. Sustainable Protein Sources . London, England : Academic Press ; 2016 . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC 13 Allès B Baudry J Méjean C , et al. . Comparison of sociodemographic and nutritional characteristics between self-reported vegetarians, vegans, and meat-eaters from the NutriNet-Santé study . Nutrients. 2017 ; 9 : E1023 . Google Scholar Crossref Search ADS PubMed WorldCat 14 Finnigan TJA. Mycoprotein: origins, production and properties. In: Phillips GO William PA , eds. Handbook of Food Proteins . London, England : Woodhead Publishing ; 2011 : 335 – 352 . Google Scholar Crossref Search ADS Google Scholar Google Preview WorldCat COPAC 15 Dunlop MV Kilroe SP Bowtell JL , et al. . Mycoprotein represents a bioavailable and insulinotropic non-animal-derived dietary protein source: a dose–response study . Br J Nutr. 2017 ; 11 : 1 – 13 . OpenURL Placeholder Text WorldCat 16 Gorissen SH Crombag JJ Senden JM , et al. . Protein content and amino acid composition of commercially available plant-based protein isolates . Amino Acids. 2018 ; 50 : 1685 – 1695 . Google Scholar Crossref Search ADS PubMed WorldCat 17 US Department of Agriculture, Agricultural Research Service. Food Data Central. Beltsville, MD: Agricultural Research Service; 2019. fdc.nal.usda.gov. Accessed July 2019. 18 Finnigan T. Mycoprotein: a healthy new protein with a low environmental impact. In: Nadathur SR Wanasundra JPD Scanlin L , eds. Sustainable Protein Sources . Cambridge, England : Academic Press ; 2012 : 305 – 326 . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC 19 Carbon Trust. Quorn, Beef and Chicken Footprints. London, England: Carbon Trust. https://www.rspo.org/acop/2016/quorn-foods-limited/M-Policies-to-PNC-waterland.pdf. Published 2014. Accessed September 2018. 20 Udall JN Lo CW Young VR , et al. . The tolerance and nutritional value of two microfungal foods in human subjects . Am J Clin Nutr. 1984 ; 40 : 285 – 292 . Google Scholar Crossref Search ADS PubMed WorldCat 21 Turnbull WH Leeds AR Edwards GD. Effect of mycoprotein on blood lipids . Am J Clin Nutr. 1990 ; 52 : 646 – 650 . Google Scholar Crossref Search ADS PubMed WorldCat 22 Turnbull WH Leeds AR Edwards DG. Mycoprotein reduces blood lipids in free-living subjects . Am J Clin Nutr. 1992 ; 55 : 415 – 419 . Google Scholar Crossref Search ADS PubMed WorldCat 23 Turnbull WH Walton J Leeds AR. Acute effects of mycoprotein on subsequent energy intake and appetite variables . Am J Clin Nutr . 1993 ; 58 : 507 – 512 . Google Scholar Crossref Search ADS PubMed WorldCat 24 Burley VJ Paul AW Blundell JE. Influence of a high-fibre food (myco-protein) on appetite: effects on satiation (within meals) and satiety (following meals) . Eur J Clin Nutr. 1993 ; 47 : 409 – 418 . Google Scholar PubMed OpenURL Placeholder Text WorldCat 25 Nakamura H Ishikawa T Akanuma M , et al. . Effect of mycoprotein intake on serum lipids of healthy subjects . Prog Med. 1994 ; 14 : 1972 – 1976 . OpenURL Placeholder Text WorldCat 26 Ishikawa T. The effect of mycoprotein intake (12g and 24g/day) over 4 weeks on serum cholesterol levels . Prog Med. 1995 ; 15 : 61 – 74 . OpenURL Placeholder Text WorldCat 27 Homma Y Nakamura H Kumagai Y , et al. . Effects of eight week ingestion of mycoprotein on plasma levels of lipids and Apo (Lipo) proteins . Prog Med. 1995 ; 15 : 183 – 195 . OpenURL Placeholder Text WorldCat 28 Turnbull WH Ward T. Mycoprotein reduces glycemia and insulinemia when taken with an oral-glucose-tolerance test . Am J Clin Nutr. 1995 ; 61 : 135 – 140 . Google Scholar Crossref Search ADS PubMed WorldCat 29 Williamson DA Geiselman PJ Lovejoy J , et al. . Effects of consuming mycoprotein, tofu or chicken upon subsequent eating behaviour, hunger and safety . Appetite. 2006 ; 46 : 41 – 48 . Google Scholar Crossref Search ADS PubMed WorldCat 30 Ruxton CHS McMillan B. The impact of mycoprotein on blood cholesterol levels: a pilot study . Br Food J. 2010 ; 112 : 1092 – 1101 . Google Scholar Crossref Search ADS WorldCat 31 Bottin JH Swann JR Cropp E , et al. . Mycoprotein reduces energy intake and postprandial insulin release without altering glucagon-like peptide-1 and peptide tyrosine-tyrosine concentrations in healthy overweight and obese adults: a randomised-controlled trial . Br J Nutr. 2016 ; 116 : 360 – 374 . Google Scholar Crossref Search ADS PubMed WorldCat 32 Owen DE Munday KA Taylor TG , et al. . The possible nature of the hypocholesterolaemic action of a mould (Fusarium) [proceedings] . Proc Nutr Soc. 1976 ; 35 : 127A – 128A . PMID: 1028059 Google Scholar PubMed OpenURL Placeholder Text WorldCat 33 Owen DE Munday KA Taylor TG , et al. . Hypocholesterolaemic action of wheat bran and a mould (Fusarium) in rats and hamsters . Proc Nutr Soc. 1975 ; 34 : 16 – 17 . PMID: 1144393 OpenURL Placeholder Text WorldCat 34 Owen DE Munday KA Taylor TG , et al. . Effects of wheat bran and a mould (Fusarium) on cholesterol excretion in rats . Proc Nutr Soc. 1976 ; 35 : 38 – 39 . PMID: 940828 OpenURL Placeholder Text WorldCat 35 Ludwig DS Pereira MA Kroenke CH , et al. . Dietary fiber, weight gain, and cardiovascular disease risk factors in young adults . JAMA. 1999 ; 282 : 1539 – 1546 . Google Scholar Crossref Search ADS PubMed WorldCat 36 Rimm EB Ascherio A Giovannucci E , et al. . Vegetable, fruit, and cereal fiber intake and risk of coronary heart disease among men . JAMA. 1996 ; 275 : 447 – 451 . Google Scholar Crossref Search ADS PubMed WorldCat 37 Wolk A Manson JE Stampfer MJ , et al. . Long-term intake of dietary fiber and decreased risk of coronary heart disease among women . JAMA. 1999 ; 281 : 1998 – 2004 . Google Scholar Crossref Search ADS PubMed WorldCat 38 Anderson JW Tietyen-Clark J. Dietary fiber: hyperlipidemia, hypertension, and coronary heart disease . Am J Gastroenterol. 1986 ; 81 : 907 – 919 . Google Scholar PubMed OpenURL Placeholder Text WorldCat 39 Jenkins DJ Kendall CW McKeown-Eyssen G , et al. . Effect of a low-glycemic index or a high-cereal fiber diet on type 2 diabetes: a randomized trial . JAMA. 2008 ; 300 : 2742 – 2753 . Google Scholar Crossref Search ADS PubMed WorldCat 40 Fukagawa NK Anderson JW Hageman G , et al. . High-carbohydrate, high-fiber diets increase peripheral insulin sensitivity in healthy young and old adults . Am J Clin Nutr. 1990 ; 52 : 524 – 528 . Google Scholar Crossref Search ADS PubMed WorldCat 41 Kirby RW Anderson JW Sieling B , et al. . Oat-bran intake selectively lowers serum low-density lipoprotein cholesterol concentrations of hypercholesterolemic men . Am J Clin Nutr. 1981 ; 34 : 824 – 829 . Google Scholar Crossref Search ADS PubMed WorldCat 42 Macfarlane GT Gibson GR Cummings JH. Comparison of fermentation reactions in different regions of the human colon . J Appl Bacteriol. 1992 ; 72 : 57 – 64 . Google Scholar PubMed OpenURL Placeholder Text WorldCat 43 Cummings JH Pomare EW Branch WJ , et al. . Short chain fatty acids in human large intestine, portal, hepatic and venous blood . Gut. 1987 ; 28 : 1221 – 1227 . Google Scholar Crossref Search ADS PubMed WorldCat 44 David LA Maurice CF Carmody RN , et al. . Diet rapidly and reproducibly alters the human gut microbiome . Nature. 2014 ; 505 : 559–563. Google Scholar Crossref Search ADS WorldCat 45 Chen WJ Anderson JW Jennings D. Propionate may mediate the hypocholesterolemic effects of certain soluble plant fibers in cholesterol-fed rats . Proc Soc Exp Biol Med. 1984 ; 175 : 215 – 218 . Google Scholar Crossref Search ADS PubMed WorldCat 46 Ge H Li X Weiszmann J , et al. . Activation of G protein-coupled receptor 43 in adipocytes leads to inhibition of lipolysis and suppression of plasma free fatty acids . Endocrinology. 2008 ; 149 : 4519 – 4526 . Google Scholar Crossref Search ADS PubMed WorldCat 47 Chambers ES Viardot A Psichas A , et al. . Effects of targeted delivery of propionate to the human colon on appetite regulation, body weight maintenance and adiposity in overweight adults . Gut. 2015 ; 64 : 1744 – 1754 . Google Scholar Crossref Search ADS PubMed WorldCat 48 Chambers ES Byrne CS Morrison DJ , et al. . Dietary supplementation with inulin-propionate ester or inulin improves insulin sensitivity in adults with overweight and obesity with distinct effects on the gut microbiota, plasma metabolome and systemic inflammatory responses: a randomised cross-over trial . Gut. 2019 ; 68 : 1430 – 1438 . Google Scholar Crossref Search ADS PubMed WorldCat 49 Tirosh A Calay ES Tuncman G , et al. . The short-chain fatty acid propionate increases glucagon and FABP4 production, impairing insulin action in mice and humans . Sci Transl Med. 2019 ; 11 : eaav0120. Google Scholar Crossref Search ADS PubMed WorldCat 50 Harris HC Edwards CA Morrison DJ. Short chain fatty acid production from mycoprotein and mycoprotein fibre in an in vitro fermentation model . Nutrients. 2019 ; 11 : E800. Google Scholar Crossref Search ADS PubMed WorldCat 51 Wolever TM Spadafora PJ Cunnane SC , et al. . Propionate inhibits incorporation of colonic [1,2-13C] acetate into plasma lipids in humans . Am J Clin Nutr. 1995 ; 61 : 1241 – 1247 . Google Scholar Crossref Search ADS PubMed WorldCat 52 Jenkins DJ Wolever TM Leeds AR , et al. . Dietary fibres, fibre analogues, and glucose tolerance: importance of viscosity . Br Med J. 1978 ; 1 : 1392 – 1394 . Google Scholar Crossref Search ADS PubMed WorldCat 53 Nazare JA Normand S Oste Triantafyllou A , et al. . Modulation of the postprandial phase by β-glucan in overweight subjects: effects on glucose and insulin kinetics . Mol Nutr Food Res. 2009 ; 53 : 361 – 369 . Google Scholar Crossref Search ADS PubMed WorldCat 54 Blackburn NA Holgate AM Read NW. Does guar gum improve post-prandial hyperglycaemia in humans by reducing small intestinal contact area? Br J Nutr. 1984 ; 52 : 197 – 204 . Google Scholar Crossref Search ADS PubMed WorldCat 55 World Health Organization . Obesity: Preventing and Managing the Global Epidemic. Geneva, Switzerland : World Health Organization ; 2000 . Technical Report Series no. 894. https://www.who.int/nutrition/publications/obesity/WHO_TRS_894/en/ 56 Hall KD Bemis T Brychta R , et al. . Calorie for calorie, dietary fat restriction results in more body fat loss than carbohydrate restriction in people with obesity . Cell Metab. 2015 ; 22 : 427 – 436 . Google Scholar Crossref Search ADS PubMed WorldCat 57 Larsen TM Dalskov SM van Baak M , et al. . Diets with high or low protein content and glycemic index for weight-loss maintenance . N Engl J Med. 2010 ; 363 : 2102 – 2113 . Google Scholar Crossref Search ADS PubMed WorldCat 58 Keim NL Stern JS Havel PJ. Relation between circulating leptin concentrations and appetite during a prolonged, moderate energy deficit in women . Am J Clin Nutr. 1998 ; 68 : 794 – 801 . Google Scholar Crossref Search ADS PubMed WorldCat 59 Weiss EP Jordan RC Frese EM , et al. . Effects of weight loss on lean mass, strength, bone, and aerobic capacity . Med Sci Sports Exerc. 2017 ; 49 : 206–217. Google Scholar PubMed OpenURL Placeholder Text WorldCat 60 Ravussin E Lillioja S Knowler WC , et al. . Reduced rate of energy expenditure as a risk factor for body-weight gain . N Engl J Med. 1988 ; 318 : 467 – 472 . Google Scholar Crossref Search ADS PubMed WorldCat 61 Westerterp-Plantenga MS Nieuwenhuizen A Tomé D , et al. . Dietary protein, weight loss, and weight maintenance . Annu Rev Nutr. 2009 ; 29 : 21 – 41 . Google Scholar Crossref Search ADS PubMed WorldCat 62 Weigle DS Breen PA Matthys CC , et al. . A high-protein diet induces sustained reductions in appetite, ad libitum caloric intake, and body weight despite compensatory changes in diurnal plasma leptin and ghrelin concentrations . Am J Clin Nutr. 2005 ; 82 : 41 – 48 . Google Scholar Crossref Search ADS PubMed WorldCat 63 Hochstenbach-Waelen A Veldhorst MA Nieuwenhuizen AG , et al. . Comparison of 2 diets with either 25% or 10% of energy as casein on energy expenditure, substrate balance, and appetite profile . Am J Clin Nutr. 2009 ; 89 : 831 – 838 . Google Scholar Crossref Search ADS PubMed WorldCat 64 Hochstenbach-Waelen A Westerterp-Plantenga MS Veldhorst MA , et al. . Single-protein casein and gelatin diets affect energy expenditure similarly but substrate balance and appetite differently in adults . J Nutr. 2009 ; 139 : 2285 – 2292 . Google Scholar Crossref Search ADS PubMed WorldCat 65 Mettler S Mitchell N Tipton KD. Increased protein intake reduces lean body mass loss during weight loss in athletes . Med Sci Sports Exerc. 2010 ; 42 : 326 – 337 . Google Scholar Crossref Search ADS PubMed WorldCat 66 Karl JP Roberts SB. Energy density, energy intake, and body weight regulation in adults . Adv Nutr. 2014 ; 5 : 835 – 850 . Google Scholar Crossref Search ADS PubMed WorldCat 67 Veldhorst MA Nieuwenhuizen AG Hochstenbach-Waelen A , et al. . A breakfast with alpha-lactalbumin, gelatin, or gelatin + TRP lowers energy intake at lunch compared with a breakfast with casein, soy, whey, or whey-GMP . Clin Nutr. 2009 ; 28 : 147 – 155 . Google Scholar Crossref Search ADS PubMed WorldCat 68 Bertenshaw EJ Lluch A Yeomans MR. Perceived thickness and creaminess modulates the short-term satiating effects of high-protein drinks . Br J Nutr. 2013 ; 110 : 578 – 586 . Google Scholar Crossref Search ADS PubMed WorldCat 69 McCrickerd K Chambers L Brunstrom JM , et al. . Subtle changes in the flavour and texture of a drink enhance expectations of satiety . Flavour. 2012 ; 1 : 20. Google Scholar Crossref Search ADS WorldCat 70 Yeomans MR Chambers L. Satiety-relevant sensory qualities enhance the satiating effects of mixed carbohydrate-protein preloads . Am J Clin Nutr. 2011 ; 94 : 1410 – 1417 . Google Scholar Crossref Search ADS PubMed WorldCat 71 Hogenkamp PS Stafleu A Mars M , et al. . Texture, not flavor, determines expected satiation of dairy products . Appetite. 2011 ; 57 : 635 – 641 . Google Scholar Crossref Search ADS PubMed WorldCat 72 Rennie MJ Edwards RH Halliday D , et al. . Muscle protein synthesis measured by stable isotope techniques in man: the effects of feeding and fasting . Clin Sci. 1982 ; 63 : 519 – 523 . Google Scholar Crossref Search ADS PubMed WorldCat 73 Moore DR Tang JE Burd NA , et al. . Differential stimulation of myofibrillar and sarcoplasmic protein synthesis with protein ingestion at rest and after resistance exercise . J Physiol. 2009 ; 587(pt 4) : 897 – 904 . Google Scholar Crossref Search ADS WorldCat 74 Tipton KD Gurkin BE Matin S , et al. . Nonessential amino acids are not necessary to stimulate net muscle protein synthesis in healthy volunteers . J Nutr Biochem. 1999 ; 10 : 89 – 95 . Google Scholar Crossref Search ADS PubMed WorldCat 75 Rieu I Balage M Sornet C , et al. . Leucine supplementation improves muscle protein synthesis in elderly men independently of hyperaminoacidaemia . J Physiol. 2006 ; 575(pt 1) : 305 – 315 . Google Scholar Crossref Search ADS WorldCat 76 Louard RJ Fryburg DA Gelfand RA , et al. . Insulin sensitivity of protein and glucose metabolism in human forearm skeletal muscle . J Clin Invest. 1992 ; 90 : 2348 – 2354 . Google Scholar Crossref Search ADS PubMed WorldCat 77 Phillips SM Tipton KD Aarsland A , et al. . Mixed muscle protein synthesis and breakdown after resistance exercise in humans . Am J Physiol. 1997 ; 273(1 pt 1) : E99 – E107 . doi: 10.1152/ajpendo.1997.273.1.E99 OpenURL Placeholder Text WorldCat 78 Wilkinson SB Phillips SM Atherton PJ , et al. . Differential effects of resistance and endurance exercise in the fed state on signalling molecule phosphorylation and protein synthesis in human muscle . J Physiol. 2008 ; 586 : 3701 – 3717 . Google Scholar Crossref Search ADS PubMed WorldCat 79 Wall BT Burd NA Franssen R , et al. . Presleep protein ingestion does not compromise the muscle protein synthetic response to protein ingested the following morning . Am J Physiol Endocrinol Metab. 2016 ; 311 : E964 – E973 . doi:10.1152/ajpendo.00325.201 Google Scholar Crossref Search ADS PubMed WorldCat 80 Tipton KD Elliott TA Cree MG , et al. . Stimulation of net muscle protein synthesis by whey protein ingestion before and after exercise . Am J Physiol Endocrinol Metab 2007 ; 292 : E71 – E76 . doi: 10.1152/ajpendo.00166.2006 Google Scholar Crossref Search ADS PubMed WorldCat 81 Cermak NM Res PT de Groot LC , et al. . Protein supplementation augments the adaptive response of skeletal muscle to resistance-type exercise training: a meta-analysis . Am J Clin Nutr. 2012 ; 96 : 1454 – 1464 . Google Scholar Crossref Search ADS PubMed WorldCat 82 Morton RW Murphy KT McKellar SR , et al. . A systematic review, meta-analysis and meta-regression of the effect of protein supplementation on resistance training-induced gains in muscle mass and strength in healthy adults . Br J Sports Med. 2018 ; 52 : 376 – 384 . doi:10.1136/bjsports-2017-097608 Google Scholar PubMed OpenURL Placeholder Text WorldCat 83 Greenhaff PL Karagounis LG Peirce N , et al. . Disassociation between the effects of amino acids and insulin on signaling, ubiquitin ligases, and protein turnover in human muscle . Am J Physiol Endocrinol Metab. 2008 ; 295 : E595 – E604 . doi:10.1152/ajpendo.90411.2008 Google Scholar Crossref Search ADS PubMed WorldCat 84 Pennings B Boirie Y Senden JM , et al. . Whey protein stimulates postprandial muscle protein accretion more effectively than do casein and casein hydrolysate in older men . Am J Clin Nutr. 2011 ; 93 : 997 – 1005 . Google Scholar Crossref Search ADS PubMed WorldCat 85 Pennings B Groen B de Lange A , et al. . Amino acid absorption and subsequent muscle protein accretion following graded intakes of whey protein in elderly men . Am J Physiol Endocrinol Metab. 2012 ; 302 : E992 – E999 . Google Scholar Crossref Search ADS PubMed WorldCat 86 Tang JE Moore DR Kujbida GW , et al. . Ingestion of whey hydrolysate, casein, or soy protein isolate: effects on mixed muscle protein synthesis at rest and following resistance exercise in young men . J Appl Physiol (1985). 2009 ; 107 : 987 – 992 . Google Scholar Crossref Search ADS PubMed WorldCat 87 Gorissen SH Horstman AM Franssen R , et al. . Ingestion of wheat protein increases in vivo muscle protein synthesis rates in healthy older men in a randomized trial . J Nutr. 2016 ; 146 : 1651 – 1659 . Google Scholar Crossref Search ADS PubMed WorldCat 88 van Vliet S Burd NA van Loon LJ. The skeletal muscle anabolic response to plant- versus animal-based protein consumption . J Nutr. 2015 ; 145 : 1981 – 1991 . Google Scholar Crossref Search ADS PubMed WorldCat 89 Mitchell CJ McGregor RA D'Souza RF , et al. . Consumption of milk protein or whey protein results in a similar increase in muscle protein synthesis in middle aged men . Nutrients. 2015 ; 7 : 8685 – 8699 . Google Scholar Crossref Search ADS PubMed WorldCat 90 Hartman JW Tang JE Wilkinson SB , et al. . Consumption of fat-free fluid milk after resistance exercise promotes greater lean mass accretion than does consumption of soy or carbohydrate in young, novice, male weightlifters . Am J Clin Nutr. 2007 ; 86 : 373 – 381 . Google Scholar Crossref Search ADS PubMed WorldCat 91 West DW Burd NA Coffey VG , et al. . Rapid aminoacidemia enhances myofibrillar protein synthesis and anabolic intramuscular signaling responses after resistance exercise . Am J Clin Nutr. 2011 ; 94 : 795 – 803 . Google Scholar Crossref Search ADS PubMed WorldCat 92 Gorissen SH Burd NA Hamer HM , et al. . Carbohydrate co-ingestion delays dietary protein digestion and absorption but does not modulate postprandial muscle protein accretion . J Clin Endocrinol Metab. 2014 ; 99 : 2250 – 2258 . Google Scholar Crossref Search ADS PubMed WorldCat 93 Gorissen SHM Burd NA Kramer IF , et al. . Co-ingesting milk fat with micellar casein does not affect postprandial protein handling in healthy older men . Clin Nutr. 2017 ; 36 : 429 – 437 . Google Scholar Crossref Search ADS PubMed WorldCat 94 Staples AW Burd NA West DW , et al. . Carbohydrate does not augment exercise-induced protein accretion versus protein alone . Med Sci Sports Exerc. 2011 ; 43 : 1154 – 1161 . Google Scholar Crossref Search ADS PubMed WorldCat 95 van Vliet S Shy EL Abou Sawan S , et al. . Consumption of whole eggs promotes greater stimulation of postexercise muscle protein synthesis than consumption of isonitrogenous amounts of egg whites in young men . Am J Clin Nutr. 2017 ; 106 : 1401 – 1412 . Google Scholar Crossref Search ADS PubMed WorldCat 96 Burd NA Beals JW Martinez IG , et al. . Food-first approach to enhance the regulation of post-exercise skeletal muscle protein synthesis and remodeling . Sports Med. 2019 ; 49 : 59 – 68 . Google Scholar Crossref Search ADS PubMed WorldCat 97 Trommelen J Betz MW van Loon LJC. The muscle protein synthetic response to meal ingestion following resistance-type exercise . Sports Med. 2019 ; 49 : 185 – 197 . Google Scholar Crossref Search ADS PubMed WorldCat 98 Beals JW Mackenzie RWA van Vliet S , et al. . Protein-rich food ingestion stimulates mitochondrial protein synthesis in sedentary young adults of different BMIs . J Clin Endocrinol Metab. 2017 ; 102 : 3415 – 3424 . Google Scholar Crossref Search ADS PubMed WorldCat 99 World Health Organization . Aging and life course. WHO website. http://www.who.int/topics/ageing. Updated 2008 . Accessed September 2018. 100 Evans WJ. What is sarcopenia? J Gerontol A Biol Sci Med Sci. 1995 ; 50(special issue) : 5 – 8 . doi:10.1093/gerona/50a.special_issue.5 Google Scholar Crossref Search ADS WorldCat 101 Volpi E Sheffield-Moore M Rasmussen BB , et al. . Basal muscle amino acid kinetics and protein synthesis in healthy young and older men . JAMA. 2001 ; 286 : 1206 – 1212 . Google Scholar Crossref Search ADS PubMed WorldCat 102 Volpi E Mittendorfer B Rasmussen BB , et al. . The response of muscle protein anabolism to combined hyperaminoacidemia and glucose-induced hyperinsulinemia is impaired in the elderly . J Clin Endocrinol Metab. 2000 ; 85 : 4481 – 4490 . Google Scholar PubMed OpenURL Placeholder Text WorldCat 103 Volpi E Mittendorfer B Wolf SE , et al. . Oral amino acids stimulate muscle protein anabolism in the elderly despite higher first-pass splanchnic extraction . Am J Physiol. 1999 ; 277 : E513 – E520 . Google Scholar PubMed OpenURL Placeholder Text WorldCat 104 Wilkes EA Selby AL Atherton PJ , et al. . Blunting of insulin inhibition of proteolysis in legs of older subjects may contribute to age-related sarcopenia . Am J Clin Nutr. 2009 ; 90 : 1343 – 1350 . Google Scholar Crossref Search ADS PubMed WorldCat 105 Cuthbertson D Smith K Babraj J , et al. . Anabolic signaling deficits underlie amino acid resistance of wasting, aging muscle . FASEB J. 2005 ; 19 : 422 – 424 . Google Scholar Crossref Search ADS PubMed WorldCat 106 Pennings B Koopman R Beelen M , et al. . Exercising before protein intake allows for greater use of dietary protein-derived amino acids for de novo muscle protein synthesis in both young and elderly men . Am J Clin Nutr. 2011 ; 93 : 322 – 331 . Google Scholar Crossref Search ADS PubMed WorldCat 107 Moore DR Churchward-Venne TA Witard O , et al. . Protein ingestion to stimulate myofibrillar protein synthesis requires greater relative protein intakes in healthy older versus younger men . J Gerontol A Biol Sci Med Sci . 2015 ; 70 : 57 – 62 . Google Scholar Crossref Search ADS PubMed WorldCat 108 Bauer J Biolo G Cederholm T , et al. . Evidence-based recommendations for optimal dietary protein intake in older people: a position paper from the PROT-AGE Study Group . J Am Med Dir Assoc. 2013 ; 14 : 542 – 559 . Google Scholar Crossref Search ADS PubMed WorldCat 109 Isanejad M Mursu J Sirola J , et al. . Association of protein intake with the change of lean mass among elderly women: The Osteoporosis Risk Factor and Prevention—Fracture Prevention Study (OSTPRE-FPS ). J Nutr Sci. 2015 ; 4 : e41. doi:10.1017/jns.2015.31 Google Scholar Crossref Search ADS PubMed WorldCat 110 Beasley JM Wertheim BC LaCroix AZ , et al. . Biomarker-calibrated protein intake and physical function in the Women's Health Initiative . J Am Geriatr Soc. 2013 ; 61 : 1863 – 1871 . Google Scholar Crossref Search ADS PubMed WorldCat Author notes * These authors contributed equally to this review. © The Author(s) 2019. Published by Oxford University Press on behalf of the International Life Sciences Institute. All rights reserved. For permissions, please e-mail: [email protected]. This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/open_access/funder_policies/chorus/standard_publication_model)

Banda,, Kondwani-Joseph;Chiu,, Hsiao-Yean;Hu, Sophia, Hueylan;Yeh,, Hsiu-Chun;Lin,, Kuan-Chia;Huang,, Hui-Chuan

doi: 10.1093/nutrit/nuz097pmid: 31995192

Abstract Context Evidence has shown that essential nutrients are highly correlated with the occurrence of esophageal cancer (EC). However, findings from observational studies on the associations between dietary carbohydrate, salt consumption, and the risk of EC remain controversial. Objective The aim of this study was to conduct a systematic review and meta-analysis to confirm the associations of dietary carbohydrate and salt consumption with EC risk. Data Source Various electronic databases (PubMed, MEDLINE, Embase, Google Scholar, Cochrane Library, Chinese Electronic Periodical Services, and China Knowledge Resource Integrated) were searched up until January 31, 2019. Data Extraction Data related to patient characteristics and study characteristics were extracted by 2 independent reviewers. The risk ratio reported as relative risk (RR) or odds ratio (OR) was extracted, and random-effects models were performed to estimate the summary risk ratio. Results In total, 26 studies were included in this analysis, of which 12 studies, including 11 case-control studies and 1 cohort study, examined dietary carbohydrates, and 18 studies, including 16 case-control studies and 2 cohort studies, examined dietary salt. The pooled OR showed that dietary carbohydrate intake was inversely related to EC risk (OR = 0.62; 95% confidence interval [CI], 0.50–0.77), but positive correlations between dietary salt intake and the risk of EC were supported by the recruited case-control studies (OR = 1.97; 95% CI, 1.50–2.61) and cohort studies (RR = 1.04; 95% CI, 1.00–1.08). Conclusions Salt is an essential nutrient for body functions and biochemical processes. Providing health education and management regarding proper use of salt in daily foods and labeling the amount of sodium in manufactured products to reduce the risk of developing EC should be more appropriately performed in the general population. carbohydrate, esophageal cancer, meta-analysis, salt INTRODUCTION Esophageal cancer (EC) is the eighth most common cancer worldwide; approximately 456000 new cases were reported in 2012, and EC caused nearly 400000 deaths.1 Based on histological classification, EC consists of 2 types: esophageal adenocarcinoma (EAC) and esophageal squamous cell carcinoma (ESCC).2 Both are notorious for their poor prognoses, because most patients are diagnosed in late stages, and progressive dysphagia and vomiting lead to poor nutrition and quality of life.3 The 5-year survival rates are still below 20% despite improvements in surgical and cancer management.4 Thus, early prevention of EC is of great importance to public health. Evidence shows that the incidence of EC has been attributed to race,2 socioeconomic status,3 lifestyle, and dietary habits5 among various populations. Previous studies found that dietary habits were highly correlated with the occurrence of EC.6,7 A healthy diet containing fruits, vegetables, and dietary fiber is helpful in preventing EC, whereas unhealthy dietary habits, such as consuming hot beverages, can cause EC.8 In addition to food categories, recent studies have found that essential nutrients, such as carbohydrates, protein, fat, and salt, are significantly correlated with the occurrence of EC.9 A meta-analysis confirmed that a high intake of dietary fat is associated with EAC risk.10 However, the effects of carbohydrates and dietary salt on the risk of EC are unclear. A high-carbohydrate diet is associated with insulin resistance and release of insulin-like growth factors that lead to obesity and, subsequently, to cancer.11 Nevertheless, some studies have proposed that a greater carbohydrate intake may reduce the level of fat intake, thereby decreasing the risk of EC.12 Available evidence from case-control and prospective studies presents inconsistent results of the effects of dietary carbohydrates on the risk of EC, with some studies reporting a positive association between carbohydrate intake and EC risk,9,13 but other studies showing a negative association.14,15 The inconsistent results may be caused by differences in the recruitment of study participants from a hospital13,14 or the community9,16 and the method of assessment of exposure, such as interviews9,14 or self-reporting.16 A systematic review conducted by Kubo et al.17 found that high consumption of carbohydrates was not associated with EAC. However, only 6 studies were included, and no statistical analysis was performed to assess the effect of carbohydrates on the risk of EC. Therefore, conducting a meta-analysis on the effects of carbohydrates on EC risk is needed. Excess salt intake is known to increase DNA synthesis and the proliferation of damaged cells in the stomach leading to development of gastric cancer.18 Available evidence from case-control and prospective cohort studies presents conflicting results, with some studies showing a positive association between dietary salt and the risk of EC19,20 and other studies denying the association.21 Significant variations have been observed in those studies, including the geographic location of Eastern or Western countries,19,22 the assessment method of self-reporting or a validated questionnaire,20,21 and the study design of case-control or cohort studies.20 Therefore, conducting a meta-analysis to explore the effects of dietary salt on the risk of EC is required. Carbohydrates and salt are essential nutrients for maintaining body functions and biochemical processes, but excess intake may be harmful because it promotes the occurrence of disease. The roles of carbohydrates and dietary salt intake in the pathogenesis of EC are controversial, and no meta-analyses have concentrated on this issue. By identifying associations of dietary carbohydrates and salt with the risk of EC, further dietary management could be addressed to prevent EC in the general population. Therefore, the aim of this study was to examine the associations of dietary carbohydrates and salt with the risk of EC by performing a systematic review and meta-analysis. METHODS The study followed guidelines recommended by PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-analyses).23 Inclusion criteria Inclusion and exclusion criteria were defined according to the PICOS (population, intervention/exposure, comparator, outcome, and study design) framework (Table 1). The inclusion criteria of studies in the meta-analysis were as follows: (1) the study design was a case-control or cohort study; (2) the outcome of interest was EC; (3) the exposures of interest were consumption of carbohydrates or salt; and (4) the presented risk was composed of the relative risk (RR), odds ratio (OR), or hazard ratio (HR) estimates with the 95% confidence interval (CI). Studies in languages other than English or Chinese were excluded. Table 1 PICOS (population, intervention/exposure, comparator, outcome, and study design) criteria for inclusion and exclusion of studies Parameter . Criteria . Population Human adults Intervention/exposure High consumption of dietary carbohydrate and salt Comparator Low consumption of dietary carbohydrate and salt Outcome Incidence of esophageal cancer Study design Observational studies (cohort and case-control study) Parameter . Criteria . Population Human adults Intervention/exposure High consumption of dietary carbohydrate and salt Comparator Low consumption of dietary carbohydrate and salt Outcome Incidence of esophageal cancer Study design Observational studies (cohort and case-control study) Open in new tab Table 1 PICOS (population, intervention/exposure, comparator, outcome, and study design) criteria for inclusion and exclusion of studies Parameter . Criteria . Population Human adults Intervention/exposure High consumption of dietary carbohydrate and salt Comparator Low consumption of dietary carbohydrate and salt Outcome Incidence of esophageal cancer Study design Observational studies (cohort and case-control study) Parameter . Criteria . Population Human adults Intervention/exposure High consumption of dietary carbohydrate and salt Comparator Low consumption of dietary carbohydrate and salt Outcome Incidence of esophageal cancer Study design Observational studies (cohort and case-control study) Open in new tab Search strategy Three authors (K.-J.B., H.-C.Y., and H.-C.H.) independently searched eligible articles according to the designated strategy. Various databases (Pubmed, Embase, MEDLINE, Google Scholar, Cochrane Library, Chinese Electronic Periodical Services, and China Knowledge Resource Integrated), from inception to January 31, 2019, were searched to include eligible articles using the following MeSH (Medical Subject Headings) terms in combination: “‘esophageal cancer’ OR ‘cancer’ OR ‘esophageal neoplasms’ OR ‘esophageal’ OR ‘neoplasms’” AND “‘carbohydrates’ OR ‘carbohydrate’ OR ‘sodium chloride’ OR ‘sodium’ OR ‘dietary sodium’ OR ‘salt’ OR ‘dietary salt.’” Furthermore, reference list of all relevant research papers and reviews were searched to identify additional eligible studies. After removing duplicates, 3 authors (K.-J.B., H.-C.Y., and H.-C.H.) independently screened the titles and abstracts to identify eligibility for inclusion. To determine the relevance of uncertain titles and abstracts, the full texts of those articles were subsequently reviewed and evaluated. Both reviewers independently assessed all abstracts and full texts of potential studies. Disagreements were resolved by discussion. Data extraction Two investigators (H.-C.Y., and H.-C.H.) independently extracted the data according to a designated collection form. The extracted data comprised the first author’s last name, year of publication, country, study design (case control or cohort), number of study participants, method of exposure assessment, instrument used for assessment, adjusted covariates, and HRs, RRs, or ORs (95% CIs). Disagreements were resolved by discussion. Assessment of study quality The internal validity of each study in the meta-analysis was appraised using the Newcastle-Ottawa scale (NOS). Nine items were assessed in each study and categorized into 3 groups: selection of study groups, comparability of the groups, and ascertainment of either the exposure or outcome of interest for case-control or cohort studies. Points were awarded for each item, with the highest being 9 points. Studies with a total of 7 points or greater were considered to be of high quality, whereas studies with less than 7 points were considered to be of poor quality.24 Statistical analysis Comprehensive Meta-Analysis 2.0 software (Biostat, Englewood, NJ, USA) was used to manage and analyze the data. The inverse variance-weighted mean of the logarithm of the RR or OR with its 95% CI was calculated to assess the effect of dietary carbohydrates or salt on the risk of EC using random- and fixed-effects models.25 Cochran's Q statistic and I2 statistic were used to examine the heterogeneity of the included studies.26 A statistically significant Cochran's Q statistic indicated that the true effect size was not the same for all studies (P < 0.10), and the I2 statistic was used to quantify heterogeneity with 25%, 50%, and 75% indicating low, moderate, and high heterogeneity, respectively.26 Because the highest and lowest categories of exposure were inconsistent in these studies, summary risk estimates were obtained by combining study-specific linear trends using random-effects models. The pooled risk estimates indicated that the association between the EC risk and high-level or low-level consumption of dietary carbohydrates and salt. Moderator analyses were performed to explore sources of variation among the studies using specified characteristics (age, gender, study location, study setting, type of dietary salt, NOS score, sample size, type of instrument, method of assessing exposure, and adjusted variables). A subgroup analysis was conducted to compare effect sizes among studies in groups of interest for categorical moderators, and a metaregression model was conducted to examine the relationship of effect sizes and variables of interest for continuous moderators.27 Moderator analyses were limited to instances in groups that were represented by at least 2 studies to ensure that sufficient data could be obtained for the analysis. Publication bias was examined using the fail-safe N test and the trim-and-fill method. The fail-safe N test estimated the number of additional studies required to render the effect size insignificant from included and additional studies combined, and a large number of estimated studies suggested that the effect size was robust, indicating no publication bias.28 The trim-and-fill method was performed to adjust for publication bias by reexamining the pooled effect size after considering the number of missing studies in the meta-analysis.29 RESULTS Literature search In total, 3169 studies were identified from the designated data sets and other articles identified through reference lists from relevant articles. In all, 2965 studies were screened after 216 duplicates were removed. Thirty-three studies were identified for full text review, excluding 2932 studies based on 667 being non-EC studies, 285 being non-dietary-nutrient studies, and 1980 not being case-control or cohort studies. After reviewing the 33 eligible studies, 7 studies were excluded based on 1 study having a similar study design with one of the included studies in the meta-analysis,30 the data of 4 studies being unavailable,31–34 and 2 studies being published in Portuguese or Polish.35,36 Ultimately, 26 articles were included for further analysis. Of these, 12 studies examined the correlation between carbohydrate consumption and EC risk, and 18 studies investigated the effect of dietary salt on the risk of EC. Five studies adopted the 2 histological types of EC (i.e., ESCC and EAC) as outcomes, 2 studies divided study participants into male and female participants, and 1 study classified dietary sodium consumption as salted meat and salted fish. Therefore, totals of 16 and 23 effect sizes were used for data analysis of dietary carbohydrates and salt, respectively. The process of study selection is shown in Figure 1.9,12–14,16,19–22,37–53 Figure 1 Open in new tabDownload slide Preferred reporting items for systematic reviews and meta-analyses (PRISMA) 2009 flow diagram. Figure 1 Open in new tabDownload slide Preferred reporting items for systematic reviews and meta-analyses (PRISMA) 2009 flow diagram. Study characteristics In total, 23 case-control studies and 3 cohort studies published between 1994 and 2018 were included. The mean subject age was 61.4 years (range, 51.3–68.0 years), and the percentage of male participants was 66.6% (range, 39.0%–100.0%). Sample sizes of these studies ranged from 143 to 494978; 9073 EC cases were included, and numbers of cases in these studies ranged from 43 to 942. Nine studies recruited study participants from hospitals, whereas 17 studies recruited from the community. Moreover, 16 studies were conducted in developing countries in South America, as well as China, India, and Iran, whereas 10 studies were conducted in developed countries including Australia, Greece, Ireland, Sweden, Taiwan, and the United States (see Table S1 in the Supporting Information online). In terms of the EC type, most studies included ESCC only (N = 13), 3 studies included only EAC, 5 studies included both ESCC and EAC, and 5 studies described only EC. Regarding the assessment of dietary exposure, 22 studies used face-to-face interviews, 3 studies used self-reporting, and 1 study used telephone interviews to complete data collection. Moreover, 16 studies adopted validated questionnaires (i.e., a food frequency questionnaire or a health habits and history questionnaire) to assess dietary exposure, and 10 studies did not specify whether the questionnaires used were validated. Regarding exposure details, most studies categorized exposure into various levels such as tertiles, quartiles, or quintiles according to the lowest to highest dietary carbohydrate and salt consumption (N = 14), and 11 studies described definite doses for the highest and lowest levels of dietary carbohydrate and salt exposure. In terms of quality assessment of the studies evaluated using the NOS, 16 studies were determined to be good quality, with NOS scores of 7 points or greater, whereas 10 studies were shown to have fair or poor quality (NOS < 7; see Table S2 in the Supporting Information online). Dietary carbohydrate consumption and the risk of esophageal cancer Case-control studies. The random-effects model found an inverse association between carbohydrate intake and EC risk among the 11 case-control studies. The pooled OR comparing the highest and lowest levels of dietary carbohydrate intake was 0.62 (95% CI, 0.50–0.77), with moderate heterogeneity among studies (Q statistic = 30.21, P = 0.00, I2 = 53.7%; Figure 2)9,12–14,16,19,42,43,45,46,53. Those participants with greater dietary carbohydrate intake had a significant reduction (38%) in EC risk. Figure 2 Open in new tabDownload slide Forest plot of the risk of esophageal cancer for the highest and lowest categories of carbohydrate consumption: case-control studies. Figure 2 Open in new tabDownload slide Forest plot of the risk of esophageal cancer for the highest and lowest categories of carbohydrate consumption: case-control studies. Cohort study. Only 1 cohort study was conducted to examine the significant association of carbohydrate consumption with the risk of EC (adjusted HR = 0.90; 95% CI, 0.53–1.51).47 Dietary sodium consumption and the risk of esophageal cancer Case-control studies. The random-effects model of the 16 included case-control studies showed a positive association between dietary sodium intake and EC risk. Compared with the lowest level of dietary sodium intake, the pooled OR of the highest level was 1.97 (95% CI, 1.49–2.61), with significant heterogeneity among the studies (Q statistic = 105.97, P < 0.001, I2 = 82.1%; Figure 3A). 12,14,19–21,37,39–41,43,44,48–52 Figure 3 Open in new tabDownload slide Forest plot of the risk of esophageal cancer for the highest and lowest categories of dietary sodium consumption. Figure 3 Open in new tabDownload slide Forest plot of the risk of esophageal cancer for the highest and lowest categories of dietary sodium consumption. Cohort studies. The random-effects model of the 2 cohort studies showed a significant positive association between dietary sodium consumption and EC risk. Compared with participants who had a lower level of dietary sodium intake, those with the highest level of dietary sodium intake had a significant risk of developing EC (HR = 1.04; 95% CI, 1.00–1.08), with no heterogeneity among the studies (Q statistic = 1.07, P = 0.585, I2 = 0%; Figure 3B). 22,38 Moderator analysis Dietary carbohydrate consumption and the risk of esophageal cancer in case-control studies. The metaregression analysis showed that effect sizes were significantly associated with age, indicating that older participants demonstrated a greater risk of developing EC (β = 0.12, P = 0.002). Moreover, a subgroup analysis indicated that studies adjusting for the variable of energy intake in the final model examining the association between dietary carbohydrate intake and risk of EC had a large OR compared with those studies that did not adjust for the variable of energy intake (OR = 0.69 vs 0.37, P = 0.008). Results indicated that those studies that did not consider energy intake as a covariate demonstrated a greater reduction in the risk of EC when comparing the highest and lowest consumption levels of dietary carbohydrates compared with studies with the covariate of energy intake (63% vs 31%; Table 2). Table 2 Moderator analysis and metaregression . Carbohydrates . Dietary salt . Variables . n . OR (95% CI) . P value . N . OR (95% CI) . P value . Age (β coefficient) 11 0.12 (0.04, 0.19) 0.002 17 −0.03 (−0.09,−0.04) 0.408 Gender (male %) (β coefficient) 15 0.01 (−0.01, 0.02) 0.262 20 −0.00 (−0.01, 0.01) 0.096 Type of EC 0.791 <0.001 ESCC 8 0.64 (0.47, 0.90) 0.012 13 2.28 (1.65, 3.15) <0.001 EAC 7 0.60 (0.45, 0.79) <0.001 2 0.95 (0.67, 1.35) 0.774 Location 0.225 0.010 Developed 11 0.66 (0.52, 0.84) 0.001 4 1.19 (0.81, 1.75) 0.379 Developing 4 0.45 (0.25, 0.80) 0.006 16 2.24 (1.68, 3.00) <0.001 NOS score 0.355 0.773 <7 7 0.56 (0.38, 0.82) 0.003 7 2.08 (1.22, 3.56) 0.007 ≥7 8 0.69 (0.54, 0.88) 0.003 13 1.90 (1.41, 2.57) <0.001 Setting 0.218 0.168 Community 10 0.57 (0.45, 0.72) <0.001 13 1.70 (1.21, 2.39) 0.002 Hospital 5 0.76 (0.51, 1.12) 0.162 7 2.50 (1.63, 3.84) <0.001 Type of instrument 0.339 Validated 14 ND 10 1.74 (1.22, 2.47) 0.002 Not validated 1 ND 10 2.27 (1.50, 3.42) <0.001 Method of assessment 0.871 Face-to-face interview 12 0.64 (0.49, 0.82) 0.001 19 ND Self-report 2 0.61 (0.36, 1.03) 0.064 1 ND Type of dietary salt 0.001 Sodium ND 6 1.24 (0.97, 1.58) 0.084 Salted food ND 14 2.46 (1.81, 3.35) <0.001 Adjustments Alcohol 0.142 0.481 Yes 11 0.67 (0.53, 0.85) 0.001 14 1.83 (1.35, 2.46) <0.001 No 4 0.45 (0.28, 0.73) 0.001 6 2.29 (1.31, 4.00) 0.003 BMI 0.641 0.182 Yes 11 0.59 (0.47, 0.74) <0.001 7 1.54 (0.94, 2.54) 0.089 No 4 0.68 (0.40, 1.15) 0.150 13 2.30 (1.69, 3.13) <0.001 Smoking 0.068 0.329 Yes 13 0.65 (0.53, 0.81) <0.001 15 1.80 (1.36, 2.37) <0.001 No 2 0.38 (0.22, 0.65) <0.001 5 2.51 (1.37, 4.60) 0.003 Energy intake 0.008 0.396 Yes 11 0.69 (0.56, 0.86) 0.001 6 1.64 (0.95, 2.85) 0.078 No 4 0.37 (0.24, 0.56) <0.001 14 2.16 (1.58, 2.95) <0.001 . Carbohydrates . Dietary salt . Variables . n . OR (95% CI) . P value . N . OR (95% CI) . P value . Age (β coefficient) 11 0.12 (0.04, 0.19) 0.002 17 −0.03 (−0.09,−0.04) 0.408 Gender (male %) (β coefficient) 15 0.01 (−0.01, 0.02) 0.262 20 −0.00 (−0.01, 0.01) 0.096 Type of EC 0.791 <0.001 ESCC 8 0.64 (0.47, 0.90) 0.012 13 2.28 (1.65, 3.15) <0.001 EAC 7 0.60 (0.45, 0.79) <0.001 2 0.95 (0.67, 1.35) 0.774 Location 0.225 0.010 Developed 11 0.66 (0.52, 0.84) 0.001 4 1.19 (0.81, 1.75) 0.379 Developing 4 0.45 (0.25, 0.80) 0.006 16 2.24 (1.68, 3.00) <0.001 NOS score 0.355 0.773 <7 7 0.56 (0.38, 0.82) 0.003 7 2.08 (1.22, 3.56) 0.007 ≥7 8 0.69 (0.54, 0.88) 0.003 13 1.90 (1.41, 2.57) <0.001 Setting 0.218 0.168 Community 10 0.57 (0.45, 0.72) <0.001 13 1.70 (1.21, 2.39) 0.002 Hospital 5 0.76 (0.51, 1.12) 0.162 7 2.50 (1.63, 3.84) <0.001 Type of instrument 0.339 Validated 14 ND 10 1.74 (1.22, 2.47) 0.002 Not validated 1 ND 10 2.27 (1.50, 3.42) <0.001 Method of assessment 0.871 Face-to-face interview 12 0.64 (0.49, 0.82) 0.001 19 ND Self-report 2 0.61 (0.36, 1.03) 0.064 1 ND Type of dietary salt 0.001 Sodium ND 6 1.24 (0.97, 1.58) 0.084 Salted food ND 14 2.46 (1.81, 3.35) <0.001 Adjustments Alcohol 0.142 0.481 Yes 11 0.67 (0.53, 0.85) 0.001 14 1.83 (1.35, 2.46) <0.001 No 4 0.45 (0.28, 0.73) 0.001 6 2.29 (1.31, 4.00) 0.003 BMI 0.641 0.182 Yes 11 0.59 (0.47, 0.74) <0.001 7 1.54 (0.94, 2.54) 0.089 No 4 0.68 (0.40, 1.15) 0.150 13 2.30 (1.69, 3.13) <0.001 Smoking 0.068 0.329 Yes 13 0.65 (0.53, 0.81) <0.001 15 1.80 (1.36, 2.37) <0.001 No 2 0.38 (0.22, 0.65) <0.001 5 2.51 (1.37, 4.60) 0.003 Energy intake 0.008 0.396 Yes 11 0.69 (0.56, 0.86) 0.001 6 1.64 (0.95, 2.85) 0.078 No 4 0.37 (0.24, 0.56) <0.001 14 2.16 (1.58, 2.95) <0.001 Abbreviations: BMI, body mass index; CI, confidence interval; EAC, esophageal adenocarcinoma; EC, esophageal cancer; ESCC, esophageal squamous cell carcinoma; ND, no data; NOS, Newcastle-Ottawa scale; OR, odds ratio. Open in new tab Table 2 Moderator analysis and metaregression . Carbohydrates . Dietary salt . Variables . n . OR (95% CI) . P value . N . OR (95% CI) . P value . Age (β coefficient) 11 0.12 (0.04, 0.19) 0.002 17 −0.03 (−0.09,−0.04) 0.408 Gender (male %) (β coefficient) 15 0.01 (−0.01, 0.02) 0.262 20 −0.00 (−0.01, 0.01) 0.096 Type of EC 0.791 <0.001 ESCC 8 0.64 (0.47, 0.90) 0.012 13 2.28 (1.65, 3.15) <0.001 EAC 7 0.60 (0.45, 0.79) <0.001 2 0.95 (0.67, 1.35) 0.774 Location 0.225 0.010 Developed 11 0.66 (0.52, 0.84) 0.001 4 1.19 (0.81, 1.75) 0.379 Developing 4 0.45 (0.25, 0.80) 0.006 16 2.24 (1.68, 3.00) <0.001 NOS score 0.355 0.773 <7 7 0.56 (0.38, 0.82) 0.003 7 2.08 (1.22, 3.56) 0.007 ≥7 8 0.69 (0.54, 0.88) 0.003 13 1.90 (1.41, 2.57) <0.001 Setting 0.218 0.168 Community 10 0.57 (0.45, 0.72) <0.001 13 1.70 (1.21, 2.39) 0.002 Hospital 5 0.76 (0.51, 1.12) 0.162 7 2.50 (1.63, 3.84) <0.001 Type of instrument 0.339 Validated 14 ND 10 1.74 (1.22, 2.47) 0.002 Not validated 1 ND 10 2.27 (1.50, 3.42) <0.001 Method of assessment 0.871 Face-to-face interview 12 0.64 (0.49, 0.82) 0.001 19 ND Self-report 2 0.61 (0.36, 1.03) 0.064 1 ND Type of dietary salt 0.001 Sodium ND 6 1.24 (0.97, 1.58) 0.084 Salted food ND 14 2.46 (1.81, 3.35) <0.001 Adjustments Alcohol 0.142 0.481 Yes 11 0.67 (0.53, 0.85) 0.001 14 1.83 (1.35, 2.46) <0.001 No 4 0.45 (0.28, 0.73) 0.001 6 2.29 (1.31, 4.00) 0.003 BMI 0.641 0.182 Yes 11 0.59 (0.47, 0.74) <0.001 7 1.54 (0.94, 2.54) 0.089 No 4 0.68 (0.40, 1.15) 0.150 13 2.30 (1.69, 3.13) <0.001 Smoking 0.068 0.329 Yes 13 0.65 (0.53, 0.81) <0.001 15 1.80 (1.36, 2.37) <0.001 No 2 0.38 (0.22, 0.65) <0.001 5 2.51 (1.37, 4.60) 0.003 Energy intake 0.008 0.396 Yes 11 0.69 (0.56, 0.86) 0.001 6 1.64 (0.95, 2.85) 0.078 No 4 0.37 (0.24, 0.56) <0.001 14 2.16 (1.58, 2.95) <0.001 . Carbohydrates . Dietary salt . Variables . n . OR (95% CI) . P value . N . OR (95% CI) . P value . Age (β coefficient) 11 0.12 (0.04, 0.19) 0.002 17 −0.03 (−0.09,−0.04) 0.408 Gender (male %) (β coefficient) 15 0.01 (−0.01, 0.02) 0.262 20 −0.00 (−0.01, 0.01) 0.096 Type of EC 0.791 <0.001 ESCC 8 0.64 (0.47, 0.90) 0.012 13 2.28 (1.65, 3.15) <0.001 EAC 7 0.60 (0.45, 0.79) <0.001 2 0.95 (0.67, 1.35) 0.774 Location 0.225 0.010 Developed 11 0.66 (0.52, 0.84) 0.001 4 1.19 (0.81, 1.75) 0.379 Developing 4 0.45 (0.25, 0.80) 0.006 16 2.24 (1.68, 3.00) <0.001 NOS score 0.355 0.773 <7 7 0.56 (0.38, 0.82) 0.003 7 2.08 (1.22, 3.56) 0.007 ≥7 8 0.69 (0.54, 0.88) 0.003 13 1.90 (1.41, 2.57) <0.001 Setting 0.218 0.168 Community 10 0.57 (0.45, 0.72) <0.001 13 1.70 (1.21, 2.39) 0.002 Hospital 5 0.76 (0.51, 1.12) 0.162 7 2.50 (1.63, 3.84) <0.001 Type of instrument 0.339 Validated 14 ND 10 1.74 (1.22, 2.47) 0.002 Not validated 1 ND 10 2.27 (1.50, 3.42) <0.001 Method of assessment 0.871 Face-to-face interview 12 0.64 (0.49, 0.82) 0.001 19 ND Self-report 2 0.61 (0.36, 1.03) 0.064 1 ND Type of dietary salt 0.001 Sodium ND 6 1.24 (0.97, 1.58) 0.084 Salted food ND 14 2.46 (1.81, 3.35) <0.001 Adjustments Alcohol 0.142 0.481 Yes 11 0.67 (0.53, 0.85) 0.001 14 1.83 (1.35, 2.46) <0.001 No 4 0.45 (0.28, 0.73) 0.001 6 2.29 (1.31, 4.00) 0.003 BMI 0.641 0.182 Yes 11 0.59 (0.47, 0.74) <0.001 7 1.54 (0.94, 2.54) 0.089 No 4 0.68 (0.40, 1.15) 0.150 13 2.30 (1.69, 3.13) <0.001 Smoking 0.068 0.329 Yes 13 0.65 (0.53, 0.81) <0.001 15 1.80 (1.36, 2.37) <0.001 No 2 0.38 (0.22, 0.65) <0.001 5 2.51 (1.37, 4.60) 0.003 Energy intake 0.008 0.396 Yes 11 0.69 (0.56, 0.86) 0.001 6 1.64 (0.95, 2.85) 0.078 No 4 0.37 (0.24, 0.56) <0.001 14 2.16 (1.58, 2.95) <0.001 Abbreviations: BMI, body mass index; CI, confidence interval; EAC, esophageal adenocarcinoma; EC, esophageal cancer; ESCC, esophageal squamous cell carcinoma; ND, no data; NOS, Newcastle-Ottawa scale; OR, odds ratio. Open in new tab Dietary sodium consumption and the risk of esophageal cancer in case-control studies. Table 2 shows that the type of EC, geographic location, and type of dietary salt were significant moderators in case-control studies that examined the association of dietary sodium intake with the risk of EC. Regarding the type of EC, studies examining only ESCC as the outcome showed that participants had a greater risk of developing EC compared with studies examining only EAC (OR = 2.28 vs 0.95, P < 0.001). Studies conducted in developing countries showed that participants had a greater risk of developing EC compared with studies conducted in developed countries (OR = 2.24 vs 1.19, P = 0.010). Regarding the type of dietary salt, studies that used salted foods as the exposure of interest showed that participants had a greater risk of developing EC compared with studies that used sodium as the exposure of interest (OR = 2.46 vs 1.24, P < 0.001). Publication bias Case-control studies. With respect to publication bias in case-control studies that measured the effect of dietary carbohydrates and salt on the risk of EC, the fail-safe N values were 154 and 699, respectively. The trim-and-fill method showed that no missing studies were required to input and adjust the pooled OR of assessing the association of carbohydrates and dietary salt with the risk of EC. Based on these results, no significant publication bias was observed. Cohort studies. As for publication bias in cohort studies measuring the effect of dietary salt on the risk of EC, the fail-safe N was 3, and the trim-and-fill method showed that the adjusted pooled OR after inputting 2 missing studies was 1.03 (95% CI, 0.99–1.07), indicating that the result may have been affected by a significant publication bias. Discussion The present meta-analysis recruiting 11 case-control studies found that a high intake of carbohydrates was associated with a significant reduction in the risk of EC compared with low consumption. However, 16 case-control studies and 2 cohort studies showed that a high intake of dietary salt was associated with a significantly increased risk of EC compared with low consumption. Previous individual studies showed inconsistent results in examining the associations of dietary carbohydrates and salt with EC risk, and few meta-analyses were conducted to clarify these issues. This meta-analysis recruited 26 studies (12 studies on dietary carbohydrates and 18 studies on dietary salt) and adopted rigorous procedures to include and review eligible studies, comprising strict inclusion criteria and a rigorous assessment of moderators and methodological quality to examine the associations of dietary carbohydrates and salt with the risk of EC. Thus, the results, which supported the effects of dietary carbohydrates and salt on the risk of EC, should be considered reliable. Because few cohort studies were found that examined the association of dietary carbohydrates with EC, the causal relationship between carbohydrates and EC should be interpreted with caution. Future studies with a cohort study design should be considered to clarify the issue. A high intake of carbohydrates was shown to have a protective effect on the risk of EC and was associated with a 37% reduction in the risk of EC compared with subjects with low intake. The majority of carbohydrates may be from whole grains. Consumption of whole grains such as wheat, brown rice, and barley, which are rich in vitamins and low in calories, can be beneficial for reducing EC.54,55 A previous study found that whole grains, which are rich in bioactive micronutrients consisting of vitamin B, minerals, and fiber, have a protective effect against cancer.56 Moreover, the bioactive micronutrients prevent cancer through their antioxidant activities, as well as preventing cell proliferation, metastasis, and formation of N-nitroso compounds.12,57 Although a high intake of carbohydrates from whole grains may be associated with a reduced risk of EC, the protective effect could not be explained by the meta-analysis because of unclear definitions of dietary carbohydrates. Therefore, future studies assessing the effects of specific types of dietary carbohydrates, such as wheat, brown rice, and barley, on the risk of EC are needed. Both case-control and cohort studies confirmed that a high intake of dietary salt increased the risk of developing EC compared with a low intake. The available evidence from systematic reviews and meta-analyses has shown that a high-salt diet is associated with an increased risk of gastric cancer.18,58,59 A high-salt diet may lead to cell proliferation and DNA synthesis of damaged cells in the esophagus, thereby causing EC.59 A low-salt diet of 5 mg/day, as recommended by the World Health Organization for an adult, might be beneficial in reducing the risk of various chronic diseases.60 Although a definite low dose of dietary salt was not suggested because of inconsistent definitions of the measured salt among these studies in the meta-analysis, future studies exploring the proper dose of dietary salt to decrease the risk of EC should be considered. Results of moderator analysis showed that an older age and the covariate of energy intake were significant moderators when adjusted for in the final model examining the association of dietary carbohydrates with EC risk. Most patients with EC are 50 years of age or older.4 Aging increases many degenerative pathologies that are characterized by debilitating loss of tissue or cellular function and lead to the occurrence of cancer.61 Therefore, early prevention of exposure to the risk factors for cancer (including EC) in midlife stages is an important public health issue. In addition, the subgroup analysis revealed that studies that did not adjust for energy intake might have overestimated the association of dietary carbohydrates with EC risk. A previous study found that high energy intake may increase the risk of cancers of the colon/rectum, prostate, and breast.62 Excess energy intake was highly correlated with obesity, which might increase the risk of developing EC.63 Thus, the importance of controlling energy intake when assessing the effect of consuming dietary carbohydrates in decreasing the risk of EC should be considered. Results showed that the location in which the study was conducted and the type of dietary salt were significant moderators explaining the heterogeneity among studies that examined the association of dietary salt with EC risk. Those studies that used salted food as the exposure and that were conducted in developing countries showed that participants had a greater risk of developing EC than their counterparts. A review article proposed that unhealthy nutrition patterns may exist among populations belonging to lower socioeconomic classes.64 Moreover, the 3 barriers of cost, accessibility, and knowledge may prohibit people in lower socioeconomic social strata from easily eating or choosing a healthy diet.65 Thus, populations in developing countries may tend to consume an inexpensive diet containing processed foods and have little knowledge that salted foods increase the risk of developing EC. Although correct estimation of sodium consumption in daily food is not easy or convenient, appropriate labeling of the amount of sodium in manufactured products could be considered an alternative approach for decreasing the risk of EC, especially in developing countries. Moreover, providing health education to promote a greater understanding of the risks of salted foods in the daily diet should be further considered for the general population. Regarding the histological type of EC, the subgroup analysis indicated that participants consuming the highest category of dietary salt had a greater risk of developing ESCC than EAC (OR = 2.28 vs 0.95, P < 0.001). The small number of studies focusing on EAC (n = 2) among the recruited studies may be a possible explanation. Another reason is that ESCC develops in the thin, flat cells that make up the outermost layer of skin, and they are mostly affected by heavy alcohol and tobacco use and irritating foods, whereas EAC occurs in the mucus-forming gland cells of the esophageal lining and is highly associated with obesity and persistent acid reflux.66 Irritating elements from salty foods may stimulate and damage the outermost layer of the esophagus, which would increase the risk of developing ESCC. Because ESCC is the most common type of EC,67 providing appropriate education and management to prevent people from consuming excess salted foods or sodium and contracting ESCC is an important public health issue. Results showed that a significant publication bias may have compromised the findings of cohort studies examining the association of dietary salt with EC risk. A possible explanation may be that only 2 such cohort studies were included. Previous studies found that salt consumption was highly related to the risk of gastric cancer.58,59 Because both case-control and cohort studies showed a positive correlation between dietary salt and EC risk, the results of the cohort studies may be reliable. More longitudinal studies examining the effects of dietary salt consumption on the risk of EC should be considered. This meta-analysis had some strengths, including a large number of studies, using strict criteria for inclusion of eligible studies, and making rigorous assessments of the methodological quality of the included studies, which increased the internal validity. Moreover, the inclusion of articles written in both English and Chinese increased the external validity of the review. However, some limitations must be considered when interpreting the findings. First, only 1 cohort study was recruited to examine the causal relationship between dietary carbohydrates and the risk of EC; thus, that result should be interpreted with caution. Second, a definite dose of dietary sodium to reduce the risk of developing EC could not be ascertained because of inconsistent definitions of the measured sodium among the recruited studies. More investigations need to be conducted to clarify the dose-response effect of dietary salt on the risk of developing EC. CONCLUSION This meta-analysis confirmed that the intake of dietary carbohydrates may have an inverse correlation with the risk of EC; however, the intake of dietary salt was positively associated with a significant increase in the risk of EC. Moreover, people who consume salted foods and live in developing countries showed a significantly higher risk of developing EC. Because salt is an essential seasoning in food preparation and plays critical roles in body functions and biochemical processes, providing health education and management regarding proper use of salt in the daily diet to reduce the risk of developing EC in the general population is required. Furthermore, appropriate labeling of the amount of sodium in manufactured products should be considered as an alternative approach to decrease the risk of EC. Acknowledgments None. Author contributions. K.-J.B. and H.-C.H. conceived the review, performed the literature search, extracted and analyzed the data, conducted the risk-of-bias assessment, and wrote the manuscript. H.-Y.C. and S.-H.H. refined the search protocol and edited the manuscript. H.-C.Y. contributed to the literature search and data extraction. K.-C.L. edited the final manuscript and provided methodological expertise. All authors contributed to reviewing the manuscript and approved the final draft for submission. Funding. This study was supported by Ministry of Science and Technology, Taiwan (MOST 107–2314-B-038–025). Declaration of interest. The authors report no conflicts of interest. Supporting information The following Supporting Information is available through the online version of this article at the publisher’s website. Table S1Characteristics of study participants in the included studies presented in alphabetical order Table S2Characteristics of study design in the included studies presented in alphabetical order References 1 Yang CS Chen X Tu S. Etiology and prevention of esophageal cancer . Gastrointest Tumors. 2016 ; 3 : 3 – 16 . Google Scholar Crossref Search ADS PubMed WorldCat 2 Abbas G Krasna M. Overview of esophageal cancer . Ann Cardiothorac Surg. 2017 ; 6 : 131 – 136 . Google Scholar Crossref Search ADS PubMed WorldCat 3 Mao W-M Zheng W-H Ling Z-Q. Epidemiologic risk factors for esophageal cancer development . Asian Pac J Cancer Prev . 2011 ; 12 : 2461 – 2466 . Google Scholar PubMed OpenURL Placeholder Text WorldCat 4 Zhang Y. Epidemiology of esophageal cancer . World J Gastroenterol . 2013 ; 19 : 5598 – 5606 . Google Scholar Crossref Search ADS PubMed WorldCat 5 Chai J Jamal MM. Esophageal malignancy: a growing concern . World J Gastroenterol . 2012 ; 18 : 6521 – 6526 . Google Scholar Crossref Search ADS PubMed WorldCat 6 Bravi F Edefonti V Randi G. Dietary patterns and the risk of esophageal cancer . Ann Oncol . 2012 ; 23 : 765 – 770 . Google Scholar Crossref Search ADS PubMed WorldCat 7 Tian D Mo SJ Han LK , et al. . Investigation of dietary factors and esophageal cancer knowledge: comparison of rural residents in high- and low-incidence areas . Sci Rep. 2018 ; 8 : 8663. Google Scholar Crossref Search ADS PubMed WorldCat 8 Munishi MO Hanisch R Mapunda O , et al. . Africa’s oesophageal cancer corridor: do hot beverages contribute? Cancer Causes Control. 2015 ; 26 : 1477 – 1486 . Google Scholar Crossref Search ADS PubMed WorldCat 9 Lagergren K Lindam A Lagergren J. Dietary proportions of carbohydrates, fat, and protein and risk of oesophageal cancer by histological type . PLoS ONE . 2013 ; 8 : e54913. Google Scholar Crossref Search ADS PubMed WorldCat 10 Du He XH Wang Z-P Chen D , et al. . Dietary fat intake and risk of esophageal carcinoma: a meta-analysis of observational studies . Oncotarget . 2017 ; 8 :99049–99056. OpenURL Placeholder Text WorldCat 11 Tasevska N Jiao L Cross AJ , et al. . Sugars in diet and risk of cancer in the NIH-AARP Diet and Health Study . Int J Cancer. 2012 ; 130 : 159 – 169 . Google Scholar Crossref Search ADS PubMed WorldCat 12 Chen H Tucker KL Graubard BI , et al. . Nutrient intakes and adenocarcinoma of the esophagus and distal stomach . Nutr Cancer . 2002 ; 42 : 33 – 40 . Google Scholar Crossref Search ADS PubMed WorldCat 13 Tzonou A Lipworth L Garidou A , et al. . Diet and risk of esophageal cancer by histologic type in a low-risk population . Int J Cancer . 1996 ; 68 : 300 – 304 . Google Scholar Crossref Search ADS PubMed WorldCat 14 De Stefani E Ronco AL Boffetta P , et al. . Nutrient intake and risk of squamous cell carcinoma of the esophagus: a case-control study in Uruguay . Nutr Cancer . 2006 ; 56 : 149 – 157 . Google Scholar Crossref Search ADS PubMed WorldCat 15 Lahmann PH Ibiebele T Webb PM , et al. . Glycemic index, glycemic load, and fiber intake and risk of adenocarcinomas and squamous cell carcinomas of the esophagus . Ann Nutr Metabol . 2013 ; 63 : 1186 . OpenURL Placeholder Text WorldCat 16 Lahmann PH Ibiebele TI Webb PM , et al. . A case-control study of glycemic index, glycemic load and dietary fiber intake and risk of adenocarcinomas and squamous cell carcinomas of the esophagus: the Australian Cancer Study . BMC Cancer . 2014 ; 14 :1–12. OpenURL Placeholder Text WorldCat 17 Kubo A Corley DA Jensen CD , et al. . Dietary factors and the risks of oesophageal adenocarcinoma and Barrett's oesophagus . Nutr Res Rev. 2010 ; 23 : 230 – 246 . Google Scholar Crossref Search ADS PubMed WorldCat 18 Ge S Feng X Shen L , et al. . Association between habitual dietary salt intake and risk of gastric cancer: a systematic review of observational studies . Gastroenterology Res Pract . 2012 ; 2012 :1–11. OpenURL Placeholder Text WorldCat 19 Mayne ST Risch HA Dubrow R , et al. . Nutrient intake and risk of subtypes of esophageal and gastric cancer . Cancer Epidemiol Biomarkers Prev . 2001 ; 10 : 1055 – 1062 . Google Scholar PubMed OpenURL Placeholder Text WorldCat 20 Song Q Wang X Yu IT , et al. . Processed food consumption and risk of esophageal squamous cell carcinoma: a case-control study in a high risk area . Cancer Sci. 2012 ; 103 : 2007 – 2011 . Google Scholar Crossref Search ADS PubMed WorldCat 21 Sehgal S Kaul S Gupta BB , et al. . Risk factors and survival analysis of the esophageal cancer in the population of Jammu, India . Indian J Cancer. 2012 ; 49 : 245 – 250 . Google Scholar Crossref Search ADS PubMed WorldCat 22 Murata A Fujino Y Pham TM , et al. . Prospective cohort study evaluating the relationship between salted food intake and gastrointestinal tract cancer mortality in Japan . Asia Pac J Clin Nutr . 2010 ; 19 : 564 – 571 . Google Scholar PubMed OpenURL Placeholder Text WorldCat 23 Moher D Liberati A Tetzlaff J , et al. . Preferred Reporting Items for Systematic Reviews and Meta-analyses: the PRISMA statement . BMJ . 2009 ; 339 :b2535. Google Scholar Crossref Search ADS PubMed WorldCat 24 Wells GA Shea B O’Connell D , et al. . The Newcastle-Ottawa Scale (NOS) for Assessing the Quality of Nonrandomised Studies in Meta-Analyses . Ottawa, ON : Ottawa Hospital Research Institute ; 2011 . http://www.ohri.ca/programs/clinical_epidemiology/oxford.asp. Accessed February 5, 2013. Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC 25 Dawson DV, Pihlstrom BL, Blanchette DR. Unerstanding and evaluating meta-analysis. J Am Dent Assoc. 2016;147:264–270. 26 Higgins JPT Thompson SG Deeks JJ , et al. . Measuring inconsistency in meta-analyses . BMJ . 2003 ; 327 : 557 – 560 . Google Scholar Crossref Search ADS PubMed WorldCat 27 Borenstein M Hedges LV Higgins JPT , et al. . Introduction to Meta-Analysis . Chichester, UK : John Wiley and Sons ; 2009 . Google Scholar Crossref Search ADS Google Scholar Google Preview WorldCat COPAC 28 Orwin RG. A fail-safe N for effect size . J Educ Statist . 1983 ; 8 : 157 – 159 . OpenURL Placeholder Text WorldCat 29 Peters JL Sutton AJ Jones DR , et al. . Performance of the trim and fill method in the presence of publication bias and between-study heterogeneity . Statist Med. 2007 ; 26 : 4544 – 4562 . Google Scholar Crossref Search ADS WorldCat 30 De Stefani E Ronco A Mendilaharsu M , et al. . Diet and risk of cancer of the upper aerodigestive tract – II . Nutr Oral Oncol . 1999 ; 35 : 22 – 26 . Google Scholar Crossref Search ADS WorldCat 31 Islami F Malekshah AF Kimiagar M , et al. . Patterns of food and nutrient consumption in northern Iran, a high-risk area for esophageal cancer . Nutr Cancer . 2009 ; 61 : 475 – 483 . Google Scholar Crossref Search ADS PubMed WorldCat 32 Onuk MD Oztopuz A Memik F. Risk factors for esophageal cancer in eastern Anatolia . Hepatogastroenterology . 2002 ; 49 : 1290 – 1292 . Google Scholar PubMed OpenURL Placeholder Text WorldCat 33 Wolfgarten E Rosendahl U Nowroth T , et al. . Coincidence of nutritional habits and esophageal cancer in Germany . Onkologie . 2001 ; 24 : 546 – 551 . Google Scholar PubMed OpenURL Placeholder Text WorldCat 34 Yu X Zhang T Zhang H , et al. . Comparison of lifestyle and living environment among high risk immigrant and low risk host residents: implications for esophageal cancer etiology . Asian Pac J Cancer Prev . 2010 ; 11 : 1827 – 1831 . Google Scholar PubMed OpenURL Placeholder Text WorldCat 35 Dietz J Pardo SH Furtado CD , et al. . [Risk factors related to esophageal cancer in Rio Grande do Sul, Brazil] . Rev Assoc Med Bras (1992). 1998 ; 44 : 269 – 272 . Google Scholar Crossref Search ADS PubMed WorldCat 36 Szumilo J. [ Epidemiology and risk factors of the esophageal squamous cell carcinoma .] Pol Merkur Lekarski . 2009 ; 26 : 82 – 85 . Google Scholar PubMed OpenURL Placeholder Text WorldCat 37 Lin S Wang X Huang C , et al. . Consumption of salted meat and its interactions with alcohol drinking and tobacco smoking on esophageal squamous-cell carcinoma . Int J Cancer. 2015 ; 137 : 582 – 589 . Google Scholar Crossref Search ADS PubMed WorldCat 38 Li WQ Park Y Wu JW , et al. . Index-based dietary patterns and risk of esophageal and gastric cancer in a large cohort study . Clin Gastroenterol Hepatol . 2013 ; 11 : 1130 – 1136.e2 . Google Scholar Crossref Search ADS PubMed WorldCat 39 Shah IA Mehta P Lone MM , et al. . CYP1A2*1F gene variant, alkaline salt tea intake and risk of esophageal squamous cell carcinoma . Nutr Cancer . 2018 ; 70 : 146 – 152 . Google Scholar Crossref Search ADS PubMed WorldCat 40 Wang JM Xu B Rao JY , et al. . Diet habits, alcohol drinking, tobacco smoking, green tea drinking, and the risk of esophageal squamous cell carcinoma in the Chinese population . Eur J Gastroenterol Hepatol . 2007 ; 19 : 171 – 176 . Google Scholar Crossref Search ADS PubMed WorldCat 41 Lu H Cai L Mu LN , et al. . Dietary mineral and trace element intake and squamous cell carcinoma of the esophagus in a Chinese population . Nutr Cancer . 2006 ; 55 : 63 – 70 . Google Scholar Crossref Search ADS PubMed WorldCat 42 Wu AH Tseng CC Hankin J , et al. . Fiber intake and risk of adenocarcinomas of the esophagus and stomach . Cancer Causes Control. 2007 ; 18 : 713 – 722 . Google Scholar Crossref Search ADS PubMed WorldCat 43 Jessri M Rashidkhani B Hajizadeh B , et al. . Macronutrients, vitamins and minerals intake and risk of esophageal squamous cell carcinoma: a case-control study in Iran . Nutr J. 2011 ; 10 : 137. Google Scholar Crossref Search ADS PubMed WorldCat 44 Gao YT McLaughlin JK Gridley G , et al. . Risk factors for esophageal cancer in Shanghai, China. II. Role of diet and nutrients . Int J Cancer. 1994 ; 58 : 197 – 202 . Google Scholar Crossref Search ADS PubMed WorldCat 45 Mulholland HG Cantwell MM Anderson LA , et al. . Glycemic index, carbohydrate and fiber intakes and risk of reflux esophagitis, Barrett's esophagus, and esophageal adenocarcinoma . Cancer Causes Control. 2009 ; 20 : 279 – 288 . Google Scholar Crossref Search ADS PubMed WorldCat 46 Sun G Wang T Huang G , et al. . Study on the dietary factors of esophageal cancer. In: Jazii FR , ed. Esophageal Cancer – Cell and Molecular Biology, Biomarkers, Nutrition and Treatment . Rijeka, Croatia : Intech ; 2012 : 127 – 152 . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC 47 Fan Y Yuan JM Wang R , et al. . Alcohol, tobacco, and diet in relation to esophageal cancer: the Shanghai Cohort Study . Nutr Cancer . 2008 ; 60 : 354 – 363 . Google Scholar Crossref Search ADS PubMed WorldCat 48 Xibib S Meilan H Moller H , et al. . Risk factors for oesophageal cancer in Linzhou, China: a case-control study . Asian Pac J Cancer Prev. 2003 ; 4 : 119 – 124 . Google Scholar PubMed OpenURL Placeholder Text WorldCat 49 Takezaki T Gao CM Wu JZ , et al. . Dietary protective and risk factors for esophageal and stomach cancers in a low‐epidemic area for stomach cancer in Jiangsu Province, China: comparison with those in a high‐epidemic area . Cancer Sci . 2001 ; 92 : 1157 – 1165 . OpenURL Placeholder Text WorldCat 50 Hung HC Huang MC Lee JM , et al. . Association between diet and esophageal cancer in Taiwan . J Gastroenterol Hepatol. 2004 ; 19 : 632 – 637 . Google Scholar Crossref Search ADS PubMed WorldCat 51 Dar NA, Bhat GA, Shah IA, et al. Salt tea consumption and esophageal cancer: a possible role of alkaline beverages in esophageal carcinogenesis . Int J Cancer . 2015 ;136:E704–E710. OpenURL Placeholder Text WorldCat 52 Castellsagué X Muñoz N De Stefani E , et al. . Influence of mate drinking, hot beverages and diet on esophageal cancer risk in South America . Int J Cancer . 2000 ; 88 : 658 – 664 . Google Scholar Crossref Search ADS PubMed WorldCat 53 Eslamian G Jessri M Hajizadeh B , et al. . Higher glycemic index and glycemic load diet is associated with increased risk of esophageal squamous cell carcinoma: a case-control study . Nutr Res . 2013 ; 33 : 719 – 725 . Google Scholar Crossref Search ADS PubMed WorldCat 54 Liu S Willett WC Manson JE , et al. . Relation between changes in intakes of dietary fiber and grain products and changes in weight and development of obesity among middle-aged women . Am J Clin Nutr . 2003 ; 78 : 920 – 927 . Google Scholar Crossref Search ADS PubMed WorldCat 55 Vaughan TL Davis S Kristal A , et al. . Obesity, alcohol, and tobacco as risk factors for cancers of the esophagus and gastric cardia: adenocarcinoma versus squamous cell carcinoma . Cancer Epidemiol Prev Biomarkers . 1995 ; 4 : 85 – 92 . OpenURL Placeholder Text WorldCat 56 Skeie G Braaten T Olsen A , et al. . Intake of whole grains and incidence of oesophageal cancer in the HELGA Cohort . Eur J Epidemiol. 2016 ; 31 : 405 – 414 . Google Scholar Crossref Search ADS PubMed WorldCat 57 Langner E Rzeski W. Dietary derived compounds in cancer chemoprevention . Contemp Oncol . 2012 ; 16 :394–400. OpenURL Placeholder Text WorldCat 58 Dias-Neto M Pintalhao M Ferreira M , et al. . Salt intake and risk of gastric intestinal metaplasia: systematic review and meta-analysis . Nutr Cancer . 2010 ; 62 : 133 – 147 . Google Scholar Crossref Search ADS PubMed WorldCat 59 Fang X Wei J He X , et al. . Landscape of dietary factors associated with risk of gastric cancer: a systematic review and dose-response meta-analysis of prospective cohort studies . Eur J Cancer . 2015 ; 51 : 2820 – 2832 . Google Scholar Crossref Search ADS PubMed WorldCat 60 World Health Organization . Diet, Nutrition and the Prevention of Chronic Diseases . Geneva, Switzerland : World Health Organization ; 2003 . WHO Technical Report Series 916. Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC 61 Campisi J. Aging, cellular senescence, and cancer . Annu Rev Physiol. 2013 ; 75 : 685 – 705 . Google Scholar Crossref Search ADS PubMed WorldCat 62 Pan SY DesMeules M. Energy intake, physical activity, energy balance, and cancer: epidemiologic evidence . Methods Mol Biol. 2009 ; 472 : 191 – 215 . Google Scholar Crossref Search ADS PubMed WorldCat 63 Corley DA Kubo A Zhao W. Abdominal obesity and the risk of esophageal and gastric cardia carcinomas . Cancer Epidemiol Biomarkers Prev . 2008 ; 17 : 352 – 358 . Google Scholar Crossref Search ADS PubMed WorldCat 64 Irala-Estevez JD Groth M Johansson L , et al. . A systematic review of socio-economic differences in food habits in Europe: consumption of fruit and vegetables . Eur J Clin Nutr. 2000 ; 54 : 706 – 714 . Google Scholar Crossref Search ADS PubMed WorldCat 65 Dibsdall LA Lambert N Bobbin RF , et al. . Low-income consumers' attitudes and behaviour towards access, availability and motivation to eat fruit and vegetables . Public Health Nutr. 2003 ; 6 : 159 – 168 . Google Scholar Crossref Search ADS PubMed WorldCat 66 Jain S Dhingra S. Pathology of esophageal cancer and Barrett's esophagus . Ann Cardiothorac Surg. 2017 ; 6 : 99 – 109 . Google Scholar Crossref Search ADS PubMed WorldCat 67 Daly JM Fry WA Little AG , et al. . Esophageal cancer: results of an American College of Surgeons Patient Care Evaluation Study . J Am Coll Surg . 2000 ; 190 : 562 – 572; discussion 572–573. Google Scholar Crossref Search ADS PubMed WorldCat © The Author(s) 2020. Published by Oxford University Press on behalf of the International Life Sciences Institute. All rights reserved. For permissions, please e-mail: [email protected]. This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/open_access/funder_policies/chorus/standard_publication_model)

Jayawardena,, Ranil;Ranasinghe,, Priyanga;Ranathunga,, Thilina;Mathangasinghe,, Yasith;Wasalathanththri,, Sudharshani;Hills, Andrew, P

doi: 10.1093/nutrit/nuz078pmid: 31841153

Abstract Context Obesity is defined as an abnormal or excessive accumulation of body fat. Traditionally, it has been assessed using a wide range of anthropometric, biochemical, and radiological measurements, with each having its advantages and disadvantages. Objective A systematic review of the literature was conducted to identify novel anthropometric measurements of obesity in adults. Data Sources Using a combination of MeSH terms, the PubMed database was searched. Data Extraction The current systematic review was conducted in accordance with the PRISMA guidelines. The data extracted from each study were (1) details of the study, (2) anthropometric parameter(s) evaluated, (3) study methods, (4) objectives of the study and/or comparisons, and (5) main findings/conclusions of the study. Data Analysis The search yielded 2472 articles, of which 66 studies were deemed eligible to be included. The literature search identified 25 novel anthropometric parameters. Data on novel anthropometric parameters were derived from 26 countries. Majority were descriptive cross-sectional studies (n = 43), while 22 were cohort studies. Age range of the study populations was 17–103 years, while sample size varied from 45 to 384 612. Conclusions The novel anthropometric parameters identified in the present study showed variable correlation with obesity and/or related metabolic risk factors. Some parameters involved complex calculations, while others were derived from traditional anthropometric measurements. Further research is required in order to determine the accuracy and precision. INTRODUCTION Obesity is defined as an abnormal or excessive accumulation of body fat that may impair health.1 Obesity and its precursor, overweight, are common problems in developed countries and becoming increasingly problematic in developing countries.2 Over the past 3 decades, the worldwide prevalence of obesity has more than doubled, and in 2014, an estimated 39% of the world’s adult population aged over 18 years were overweight and 13% were obese.1 Excess body fat has been shown to be deleterious for multiple organ systems, through thrombogenic, atherogenic, oncogenic, hemodynamic, and neurohumoral mechanisms.3–7 Obesity is the central and causal component in the pathogenesis of numerous diseases, including type 2 diabetes mellitus (T2DM), cardiovascular disease, and several cancers.8 However, obesity is preventable and early identification plays a key role in the likelihood of overcoming the condition and associated metabolic complications. Traditionally, overweight and obesity have been assessed on the basis of an excess of body weight, most commonly relative to height, with the assumption that excess body fat is recognized to be present at higher levels of body weight. However, many bigger (heavier) individuals may be classified as overweight (or obese) based on high levels of muscularity – that is, even in the absence of excess adiposity and/or associated metabolic risks.9 In short, body weight is not a measure of body composition and does not differentiate between the major components of body composition: fat mass and fat-free mass. The amount of body fat in different regions of the body also varies considerably between individuals and is a major factor in determining health risk. Many studies have demonstrated that central adiposity is associated with greater risk of metabolic complications.10–12 In the absence of consensus regarding the optimal gold standard technique to assess body fat in vivo, numerous proxy techniques have been used to estimate both body fat and its distribution. A wide range of anthropometric, biochemical, and radiological measurement approaches have been adopted. For example, the deuterium dilution technique is an example of a reference method used to assess total body water and subsequently estimate total fat mass based on a 2-compartment model.13 Magnetic resonance imaging and computed tomography are considered gold standard approaches for determining subcutaneous abdominal adipose tissue and intra-abdominal adipose tissue.14 However, such biochemical and radiological approaches cannot be routinely used in clinical and primary care settings to assess adiposity. Accordingly, various anthropometric measurements have been employed to assess overweight and obesity within a clinical setting. The body mass index (BMI), a ratio between stature and body weight, is the most widely used anthropometric measure to define obesity in adults.15–17 Despite the derivation of ethnic-specific cutoffs, for example lower cutoffs for Asians, the use of BMI still leads to high false-positive rates.15 Importantly, BMI does not distinguish between body fat and lean mass and therefore overestimates fatness among those who are muscular.16,18,19 Other widely used anthropometric indices of central obesity are waist circumference (WC), waist to hip ratio (WHR), and waist to height ratio (WHtR).20,21 Each index confers both advantages and disadvantages, and presently no anthropometric measurement for central adiposity satisfies the criteria of being accurate, precise, accessible, and widely acceptable.22 Accordingly, the scientific literature has repeatedly referenced the need for future studies to determine more precise, simple, and cost-effective measures for assessing obesity.15,16 The current systematic review was conducted in accordance with the PRISMA guidelines (Table S1 in the Supporting Information online) and aims to systematically evaluate the literature, and identify, compare, and contrast novel anthropometric measures to assess overweight and obesity in adults. METHODS A systematic review of published studies reporting novel anthropometric tools to define obesity among adults was undertaken in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines.23 Search strategy A comprehensive literature search was conducted in a stepwise process for studies published before December 31, 2017. During the first stage, the PubMed database was searched using the following MeSH (medical subject heading) terms: “obesity,” “overweight,” and “adiposity,” and combined with the following MeSH subheadings: “body weight and measures/diagnosis,” “body weight and measures/methods,” “anthropometry/diagnosis,” and “anthropometry/methods.” Search limits were species (“humans”), language (“English”), publication type (“journal articles”), and age (“19+ yr”). In the second stage, the total hits obtained from searching the database were screened for suitability by reading the article “title” and “abstract.” Subsequently, the filtered articles were further screened by reading the individual manuscripts, and those not satisfying inclusion criteria (described below) were excluded. This search process was conducted independently by 2 reviewers (P.R. and R.J.) and the final group of articles to be included in the review was determined through an iterative consensus process. The search strategy is summarized in Figure 1. Figure 1 Open in new tabDownload slide Flow diagram of the literature search process. Figure 1 Open in new tabDownload slide Flow diagram of the literature search process. Inclusion and exclusion criteria A study was considered eligible for data extraction if it described 1 or more novel anthropometric tools to define obesity in adults (age ≥18 y). Studies evaluating already well-established measures of obesity, such as BMI, WC, WHR, and WHtR, were excluded. In addition, conference proceedings, editorials, opinions/commentaries, and book chapters/book reviews were excluded. Data extraction and analysis Data were extracted from the included studies by one reviewer (Y.M.) using a standardized form and checked for accuracy by a second reviewer (S.W.). The data extracted from each study were (1) details of the study (lead author, country), (2) anthropometric parameter(s) evaluated in the study and its details (definition, cutoff, etc.), (3) study methods (sample size, male:female ratio, age group, study design), (4) objectives of the study and/or comparisons, and (5) main findings/conclusions of the study. Any discrepancies in the data extracted in this manner were rechecked and resolved by discussion, while a third reviewer (R.J.) was also involved where necessary. Data not presented in the published manuscript were obtained by contacting the corresponding author, or where possible calculated from the available data. When there were several studies describing the same anthropometric index, only the study with the largest sample and incorporating both males and females was selected. Studies involving novel anthropometric parameters are described in Table 1.12,24–42 Table 1 Novel anthropometric parameters Author [Ref]; Country . Indices . Study population (study design; sample size;male:female ratio; age group) . Measurement (method; cutoff) . Comparison/objective . Findings . Abulmeaty et al24 Saudi Arabia HtWrR Descriptive cross-sectional; n = 390; 167:223; 18–50 y HtWrR = Ht (cm) / WC (cm) Cutoff: NR To investigate the prediction of long-term cardiometabolic risk using anthropometric parameters HtWrRdid not discriminate the risk in either sex Al-Gindan et al25 USA MCC; MTC Case-control cases: n = 416 (194: 222) 18–88 y Controls: n = 204 (94:110) 18–86 y MCC – maximum girth of calf MTC – measured at the midpoint between the inguinal crease and the proximal border of the patella Cutoff: NR To validate new anthropometric equations to estimate MRI-measured TATM/ TATFM/total body fat and compare these with existing prediction equations using older methods New anthropometric measurement–based equations correlated better with MRI-measured TATM and TATFM than existing equations Carlsson et al26 Sweden SAD; SADHR Longitudinal cohort study; n = 3741; 1751:1990; 60 y SAD = distance between table and the top of the body at the level of the iliac crest SADHR= SAD (cm) / Ht (cm) Cutoff: NR To compare anthropometric measures (BMI, WHR, WC, SAD, WHHtR, WHtR, and SADHR ) in their ability to predict ischemic CVD BMI, WC, and WHtR were weaker predictors than SAD and SADHR in predicting ischemic CVD Chagas et al27 Brazil AbCHR Descriptive cross-sectional study; n = 337; 213:124; 23–89 y AbCHR = AbC (cm) / HC (cm) AbC measured 1 cm above the iliac crest Cutoff: NR To evaluate the association of different anthropometric parameters with the coronary atherosclerotic burden AbCHR was not an independent risk factor for coronary atherosclerotic burden Chang et al28 China BRI Descriptive cross-sectional study; n = 11 345; 5253:6092; ≥ 35 y BRI = 364.2–365.5 × (1– [WC/2π]2 / [0.5 × Ht]2) Cutoff: 0.66 (male) and 0.67 (female) To assess the capacity of BRI to identify subjects with T2DM and determine whether it is superior to BMI, WC, and WHtR BRI was not superior to BMI, WC, or WHtR for predicting the presence of T2DM Chung et al29 Korea ABSI; LBSIZ Cohort study; n = 54 893; 23 769: 31 124; 20–103 y ABSI = WC (m) / (BMI2/3 × Ht1/2) LBSI = log(ABSI) LBSIZ = (LBSI–LBSI[mean])/LBSI (SD) Cutoff: ABSI = 0.68; LBSIZ = 1.32 (male) and 1.86 (female) To assess sensitivity of ABSI and LBSIZ based on BF% and its ability to predict hypertension and impaired health-related quality of life LBSIZ is a measure of abdominal obesity that can predict hypertension and impaired health-related quality of life, and was more generalizable than ABSI Fu et al30 China BAI Descriptive cross-sectional study; n = 4868; 2323:2545; 18–96 y BAI = (HC [cm] / Ht [cm]1.5) − 18 Cutoff: (for ≥1 cardiometabolic abnormality): BAI = 26.39 (male) and 31.29 (female) To correlate new anthropometric indices and cardiometabolic abnormalities BAI was significantly associated with cardiometabolic abnormalities Haghighatdoost et al31 Iran CUN-BAE Descriptive cross-sectional study; n = 9555; 4777:4778; Mean age 38.7±15.5 y CUN-BAE formula for BF% = –44.988 + (0.503 × age) + (10.689 × sex) + (3.172 × BMI) − (0.026 × BMI2) + (0.181 × BMI × sex) − (0.02 × BMI × age) – (0.005 × BMI2 × sex) + (0.00021 × BMI2 × age) Cutoff: NR To compare the predictive power of ABSI, BMI, WHtR, and CUN-BAE for MetS and cardiovascular disease risks CUN-BAE was a good predictor of body fat percentage; however, it was not the best anthropometric predictor of cardiovascular risks Jung et al32 Korea TC Descriptive cross-sectional study; n = 315 628; 199 423:116 205 30–79 y TC: measured immediately below the gluteal fold in left thigh Cutoff: NR To study the association between TC and T2DM, in relation to age and BMI Small TC was associated with T2DM, and this association was stronger among participants with a BMI <25 kg/m2. TC is a possible T2DM marker in lean populations Kaneko et al33 Japan WHI Descriptive cross-sectional study; n = 695; 433:262; 45–74 y WHI = WC (cm) / (Ht [m] × Ht [m]) Cutoff: ≥ 35 To determine the ability of WHI as an index of abdominal obesity to predict the risk of colon cancer in Japanese persons WHI is an efficient predictor (more effective than WC and BMI) of risk of colon cancer in females Katulanda et al34 Sri Lanka XUD Descriptive cross-sectional study; n = 4485; 1772:2713; ≥18 y XUD: distance between the lower border of xiphisternum and the center of umbilicus at the end of normal expiration; Cutoff: NR Relationship between obesity-associated risk factors for CVD and XUD, WC, BMI, and WHR XUD is significantly, but weakly, associated with obesity-associated risk factors for CVD, but the association was weaker than that of BMI, WC, and WHR with obesity-associated risk factors for CVD Kommuri et al35 USA C-index Longitudinal cohort study; n = 6814; 3213:3601; 45–84 y C-index = WC (m) / (0.109 × ✓ [Wt (kg) / Ht (m)]) Cutoff: NR To explore the association between various anthropometric measures and markers of subclinical atherosclerosis C-index was associated with subclinical atherosclerosis compared to other measures. However, the association was weaker than other anthropometric parameters Krakauer and Krakauer36 USA HI; ARI Cohort study; n = 16 034; 7696:8338; Mean age 43 y HI = HC of a given person normalized to a standard height and weight ARI = the sum of function values for each individual’s combination of anthropometric index scores, denoting natural logarithm of the combined estimated hazard from 4 independent indices: HC, BMI, ABSI, and HI Cutoff: NR To evaluate HI and ARI as predictors of mortality risk HI and ARI were consistent predictors of mortality hazard Li et al37 USA WTR Descriptive cross- sectional study; n = 6277; 2994:3283; ≥20 y WTR = WC (cm) / TC (cm) (TC = circumference of the mid-thigh, perpendicular to the long axis of the thigh) Cutoff: NR To examine whether WTR performed better than WHtR, WHR, WC, and BMIin relation to T2DM among US adults For the association with T2DM, WTR performed better than the other 4 indices in men, and WTR performed similarly to WHtR, WHR, and WC, but better than BMI, in women Motamed et al38 Iran AVI Descriptive cross- sectional study; n = 3201; 1826:1375; 40–79 y AVI= (2 × WC2 + 0.7 [WC – HC2]) / 1000 Cutoffs: NR To determine the performance of WC, WHR, WHtR, C-index, and AVI to predict risk of 10-y CVD events AVI had the lowest discriminatory accuracy to predict 10-y CV events, while C-index had the highest accuracy Ononamadu et al39 Nigeria PI Descriptive cross-sectional study; n=912; 436:476; 17–79 y PI = Wt (kg) / (Ht [cm])3 Cutoffs: hypertension and pre-hypertension: PI = 14.45, 13.69 (female); 16.38, 17.65 (male) To compare anthropometric indices of obesity as correlates and potential predictors of risk of hypertension and pre-hypertension BMI, WHtR, WC, and PI were the best predictors of hypertension risk, and BMI, WC, and PI of pre-hypertension risk Rahman and Adjeroh40 USA SBSI Longitudinal cohort study; n = 11 808; Sex: NR; 18–85 y SBSI = (Ht7/4 × WC5/6) / (BSA × VTC) Cutoff: NR To evaluate SBSI as a predictor of all-cause mortality SBSI is generally linear with age and increases with increasing mortality, when compared with other popular anthropometric indices of body shape Song et al12 Finland WHHR Cohort study; n = 46 651; 24 686:21 965; 24–99 y WHHR = WC (m) / (HC [m] × Ht [m]) Cutoffs: NR To compare the ability of anthropometric indicators of obesity to predict CVD mortality WHHtR was a stronger predictor of CVD mortality than BMI Vogel et al41 Brazil AFA Cohort study; n = 116; 72:44; 30–85 y AFA = (AC * TSF/2) – (3.14*[TSF]2/4) Cutoffs: 90th percentile (obesity) To evaluate the association between obesity indices and mortality among patients with ischemic heart failure AFA did not predict survival among heart failure patients Zhou et al42 China NC Descriptive cross-sectional study; n = 4201; 2508:1693; 20–85 y NC = measured with head erect and eyes facing forward, horizontally at the upper laryngeal bulge Cutoffs: males ≥37 cm, females ≥33 cm To investigate whether NC independently contributes to the prediction of cardiometabolic risks over BMI, WC, and WHR in Chinese individuals NC independently contributed to the prediction of cardiometabolic risks in Chinese individuals Author [Ref]; Country . Indices . Study population (study design; sample size;male:female ratio; age group) . Measurement (method; cutoff) . Comparison/objective . Findings . Abulmeaty et al24 Saudi Arabia HtWrR Descriptive cross-sectional; n = 390; 167:223; 18–50 y HtWrR = Ht (cm) / WC (cm) Cutoff: NR To investigate the prediction of long-term cardiometabolic risk using anthropometric parameters HtWrRdid not discriminate the risk in either sex Al-Gindan et al25 USA MCC; MTC Case-control cases: n = 416 (194: 222) 18–88 y Controls: n = 204 (94:110) 18–86 y MCC – maximum girth of calf MTC – measured at the midpoint between the inguinal crease and the proximal border of the patella Cutoff: NR To validate new anthropometric equations to estimate MRI-measured TATM/ TATFM/total body fat and compare these with existing prediction equations using older methods New anthropometric measurement–based equations correlated better with MRI-measured TATM and TATFM than existing equations Carlsson et al26 Sweden SAD; SADHR Longitudinal cohort study; n = 3741; 1751:1990; 60 y SAD = distance between table and the top of the body at the level of the iliac crest SADHR= SAD (cm) / Ht (cm) Cutoff: NR To compare anthropometric measures (BMI, WHR, WC, SAD, WHHtR, WHtR, and SADHR ) in their ability to predict ischemic CVD BMI, WC, and WHtR were weaker predictors than SAD and SADHR in predicting ischemic CVD Chagas et al27 Brazil AbCHR Descriptive cross-sectional study; n = 337; 213:124; 23–89 y AbCHR = AbC (cm) / HC (cm) AbC measured 1 cm above the iliac crest Cutoff: NR To evaluate the association of different anthropometric parameters with the coronary atherosclerotic burden AbCHR was not an independent risk factor for coronary atherosclerotic burden Chang et al28 China BRI Descriptive cross-sectional study; n = 11 345; 5253:6092; ≥ 35 y BRI = 364.2–365.5 × (1– [WC/2π]2 / [0.5 × Ht]2) Cutoff: 0.66 (male) and 0.67 (female) To assess the capacity of BRI to identify subjects with T2DM and determine whether it is superior to BMI, WC, and WHtR BRI was not superior to BMI, WC, or WHtR for predicting the presence of T2DM Chung et al29 Korea ABSI; LBSIZ Cohort study; n = 54 893; 23 769: 31 124; 20–103 y ABSI = WC (m) / (BMI2/3 × Ht1/2) LBSI = log(ABSI) LBSIZ = (LBSI–LBSI[mean])/LBSI (SD) Cutoff: ABSI = 0.68; LBSIZ = 1.32 (male) and 1.86 (female) To assess sensitivity of ABSI and LBSIZ based on BF% and its ability to predict hypertension and impaired health-related quality of life LBSIZ is a measure of abdominal obesity that can predict hypertension and impaired health-related quality of life, and was more generalizable than ABSI Fu et al30 China BAI Descriptive cross-sectional study; n = 4868; 2323:2545; 18–96 y BAI = (HC [cm] / Ht [cm]1.5) − 18 Cutoff: (for ≥1 cardiometabolic abnormality): BAI = 26.39 (male) and 31.29 (female) To correlate new anthropometric indices and cardiometabolic abnormalities BAI was significantly associated with cardiometabolic abnormalities Haghighatdoost et al31 Iran CUN-BAE Descriptive cross-sectional study; n = 9555; 4777:4778; Mean age 38.7±15.5 y CUN-BAE formula for BF% = –44.988 + (0.503 × age) + (10.689 × sex) + (3.172 × BMI) − (0.026 × BMI2) + (0.181 × BMI × sex) − (0.02 × BMI × age) – (0.005 × BMI2 × sex) + (0.00021 × BMI2 × age) Cutoff: NR To compare the predictive power of ABSI, BMI, WHtR, and CUN-BAE for MetS and cardiovascular disease risks CUN-BAE was a good predictor of body fat percentage; however, it was not the best anthropometric predictor of cardiovascular risks Jung et al32 Korea TC Descriptive cross-sectional study; n = 315 628; 199 423:116 205 30–79 y TC: measured immediately below the gluteal fold in left thigh Cutoff: NR To study the association between TC and T2DM, in relation to age and BMI Small TC was associated with T2DM, and this association was stronger among participants with a BMI <25 kg/m2. TC is a possible T2DM marker in lean populations Kaneko et al33 Japan WHI Descriptive cross-sectional study; n = 695; 433:262; 45–74 y WHI = WC (cm) / (Ht [m] × Ht [m]) Cutoff: ≥ 35 To determine the ability of WHI as an index of abdominal obesity to predict the risk of colon cancer in Japanese persons WHI is an efficient predictor (more effective than WC and BMI) of risk of colon cancer in females Katulanda et al34 Sri Lanka XUD Descriptive cross-sectional study; n = 4485; 1772:2713; ≥18 y XUD: distance between the lower border of xiphisternum and the center of umbilicus at the end of normal expiration; Cutoff: NR Relationship between obesity-associated risk factors for CVD and XUD, WC, BMI, and WHR XUD is significantly, but weakly, associated with obesity-associated risk factors for CVD, but the association was weaker than that of BMI, WC, and WHR with obesity-associated risk factors for CVD Kommuri et al35 USA C-index Longitudinal cohort study; n = 6814; 3213:3601; 45–84 y C-index = WC (m) / (0.109 × ✓ [Wt (kg) / Ht (m)]) Cutoff: NR To explore the association between various anthropometric measures and markers of subclinical atherosclerosis C-index was associated with subclinical atherosclerosis compared to other measures. However, the association was weaker than other anthropometric parameters Krakauer and Krakauer36 USA HI; ARI Cohort study; n = 16 034; 7696:8338; Mean age 43 y HI = HC of a given person normalized to a standard height and weight ARI = the sum of function values for each individual’s combination of anthropometric index scores, denoting natural logarithm of the combined estimated hazard from 4 independent indices: HC, BMI, ABSI, and HI Cutoff: NR To evaluate HI and ARI as predictors of mortality risk HI and ARI were consistent predictors of mortality hazard Li et al37 USA WTR Descriptive cross- sectional study; n = 6277; 2994:3283; ≥20 y WTR = WC (cm) / TC (cm) (TC = circumference of the mid-thigh, perpendicular to the long axis of the thigh) Cutoff: NR To examine whether WTR performed better than WHtR, WHR, WC, and BMIin relation to T2DM among US adults For the association with T2DM, WTR performed better than the other 4 indices in men, and WTR performed similarly to WHtR, WHR, and WC, but better than BMI, in women Motamed et al38 Iran AVI Descriptive cross- sectional study; n = 3201; 1826:1375; 40–79 y AVI= (2 × WC2 + 0.7 [WC – HC2]) / 1000 Cutoffs: NR To determine the performance of WC, WHR, WHtR, C-index, and AVI to predict risk of 10-y CVD events AVI had the lowest discriminatory accuracy to predict 10-y CV events, while C-index had the highest accuracy Ononamadu et al39 Nigeria PI Descriptive cross-sectional study; n=912; 436:476; 17–79 y PI = Wt (kg) / (Ht [cm])3 Cutoffs: hypertension and pre-hypertension: PI = 14.45, 13.69 (female); 16.38, 17.65 (male) To compare anthropometric indices of obesity as correlates and potential predictors of risk of hypertension and pre-hypertension BMI, WHtR, WC, and PI were the best predictors of hypertension risk, and BMI, WC, and PI of pre-hypertension risk Rahman and Adjeroh40 USA SBSI Longitudinal cohort study; n = 11 808; Sex: NR; 18–85 y SBSI = (Ht7/4 × WC5/6) / (BSA × VTC) Cutoff: NR To evaluate SBSI as a predictor of all-cause mortality SBSI is generally linear with age and increases with increasing mortality, when compared with other popular anthropometric indices of body shape Song et al12 Finland WHHR Cohort study; n = 46 651; 24 686:21 965; 24–99 y WHHR = WC (m) / (HC [m] × Ht [m]) Cutoffs: NR To compare the ability of anthropometric indicators of obesity to predict CVD mortality WHHtR was a stronger predictor of CVD mortality than BMI Vogel et al41 Brazil AFA Cohort study; n = 116; 72:44; 30–85 y AFA = (AC * TSF/2) – (3.14*[TSF]2/4) Cutoffs: 90th percentile (obesity) To evaluate the association between obesity indices and mortality among patients with ischemic heart failure AFA did not predict survival among heart failure patients Zhou et al42 China NC Descriptive cross-sectional study; n = 4201; 2508:1693; 20–85 y NC = measured with head erect and eyes facing forward, horizontally at the upper laryngeal bulge Cutoffs: males ≥37 cm, females ≥33 cm To investigate whether NC independently contributes to the prediction of cardiometabolic risks over BMI, WC, and WHR in Chinese individuals NC independently contributed to the prediction of cardiometabolic risks in Chinese individuals AbC, abdominal circumference, AbCHR, abdominal circumference to hip ratio; ABSI, a body shape index; ABSIz, a body shape index z-score AC, arm circumference; AFA, arm fat area; AH, abdominal height; ARI, anthropometric risk index; AUC, area under the curve; AVI, abdominal volume index; BAI, body adiposity index; BF, body fat; BMI, body mass index; BRI, body roundness index; BSA, body surface area; C-index, conicity index; CUN-BAE, Clinica Universidad de Navarra-Body Adiposity Estimator; CV, cardiovascular; CVD, cardiovascular disease; F, female; HC, hip circumference; HCNC, height-corrected neck circumference; HI, hip index; Ht, height; HtWrR, height to wrist ratio; LBSIZ, log-transformed body shape index; M, male; MAC, mid-arm circumference; MAMA, mid-arm muscular area; MCC, mid-calf circumference; MetS, metabolic syndrome; MRI, magnetic resonance imaging; MTC, mid-thigh circumference; NC, neck circumference; NR, not reported; OSA, obstructive sleep apnea; PCOS, polycystic ovarian syndrome; PI, ponderal index; SAD, sagittal abdominal diameter; SADHR, sagittal abdominal diameter to height ratio; SBSI, surface-based body shape index; TATFM, total adipose tissue fat mass; TATM, total adipose tissue mass; TC, thigh circumference; T2DM, type 2 diabetes mellitus; TSF, triceps skinfold thickness; VAT, visceral adipose tissue; VTC, vertical trunk circumference; WC, waist circumference; WHHR, waist circumference to hip circumference to height ratio; WHI, waist to height index; WHR, waist to hip ratio; WHtR, waist to height ratio; Wt, weight; WTR, waist circumference to thigh circumference ratio; XUD, xiphisternum to umbilicus distance. Open in new tab Table 1 Novel anthropometric parameters Author [Ref]; Country . Indices . Study population (study design; sample size;male:female ratio; age group) . Measurement (method; cutoff) . Comparison/objective . Findings . Abulmeaty et al24 Saudi Arabia HtWrR Descriptive cross-sectional; n = 390; 167:223; 18–50 y HtWrR = Ht (cm) / WC (cm) Cutoff: NR To investigate the prediction of long-term cardiometabolic risk using anthropometric parameters HtWrRdid not discriminate the risk in either sex Al-Gindan et al25 USA MCC; MTC Case-control cases: n = 416 (194: 222) 18–88 y Controls: n = 204 (94:110) 18–86 y MCC – maximum girth of calf MTC – measured at the midpoint between the inguinal crease and the proximal border of the patella Cutoff: NR To validate new anthropometric equations to estimate MRI-measured TATM/ TATFM/total body fat and compare these with existing prediction equations using older methods New anthropometric measurement–based equations correlated better with MRI-measured TATM and TATFM than existing equations Carlsson et al26 Sweden SAD; SADHR Longitudinal cohort study; n = 3741; 1751:1990; 60 y SAD = distance between table and the top of the body at the level of the iliac crest SADHR= SAD (cm) / Ht (cm) Cutoff: NR To compare anthropometric measures (BMI, WHR, WC, SAD, WHHtR, WHtR, and SADHR ) in their ability to predict ischemic CVD BMI, WC, and WHtR were weaker predictors than SAD and SADHR in predicting ischemic CVD Chagas et al27 Brazil AbCHR Descriptive cross-sectional study; n = 337; 213:124; 23–89 y AbCHR = AbC (cm) / HC (cm) AbC measured 1 cm above the iliac crest Cutoff: NR To evaluate the association of different anthropometric parameters with the coronary atherosclerotic burden AbCHR was not an independent risk factor for coronary atherosclerotic burden Chang et al28 China BRI Descriptive cross-sectional study; n = 11 345; 5253:6092; ≥ 35 y BRI = 364.2–365.5 × (1– [WC/2π]2 / [0.5 × Ht]2) Cutoff: 0.66 (male) and 0.67 (female) To assess the capacity of BRI to identify subjects with T2DM and determine whether it is superior to BMI, WC, and WHtR BRI was not superior to BMI, WC, or WHtR for predicting the presence of T2DM Chung et al29 Korea ABSI; LBSIZ Cohort study; n = 54 893; 23 769: 31 124; 20–103 y ABSI = WC (m) / (BMI2/3 × Ht1/2) LBSI = log(ABSI) LBSIZ = (LBSI–LBSI[mean])/LBSI (SD) Cutoff: ABSI = 0.68; LBSIZ = 1.32 (male) and 1.86 (female) To assess sensitivity of ABSI and LBSIZ based on BF% and its ability to predict hypertension and impaired health-related quality of life LBSIZ is a measure of abdominal obesity that can predict hypertension and impaired health-related quality of life, and was more generalizable than ABSI Fu et al30 China BAI Descriptive cross-sectional study; n = 4868; 2323:2545; 18–96 y BAI = (HC [cm] / Ht [cm]1.5) − 18 Cutoff: (for ≥1 cardiometabolic abnormality): BAI = 26.39 (male) and 31.29 (female) To correlate new anthropometric indices and cardiometabolic abnormalities BAI was significantly associated with cardiometabolic abnormalities Haghighatdoost et al31 Iran CUN-BAE Descriptive cross-sectional study; n = 9555; 4777:4778; Mean age 38.7±15.5 y CUN-BAE formula for BF% = –44.988 + (0.503 × age) + (10.689 × sex) + (3.172 × BMI) − (0.026 × BMI2) + (0.181 × BMI × sex) − (0.02 × BMI × age) – (0.005 × BMI2 × sex) + (0.00021 × BMI2 × age) Cutoff: NR To compare the predictive power of ABSI, BMI, WHtR, and CUN-BAE for MetS and cardiovascular disease risks CUN-BAE was a good predictor of body fat percentage; however, it was not the best anthropometric predictor of cardiovascular risks Jung et al32 Korea TC Descriptive cross-sectional study; n = 315 628; 199 423:116 205 30–79 y TC: measured immediately below the gluteal fold in left thigh Cutoff: NR To study the association between TC and T2DM, in relation to age and BMI Small TC was associated with T2DM, and this association was stronger among participants with a BMI <25 kg/m2. TC is a possible T2DM marker in lean populations Kaneko et al33 Japan WHI Descriptive cross-sectional study; n = 695; 433:262; 45–74 y WHI = WC (cm) / (Ht [m] × Ht [m]) Cutoff: ≥ 35 To determine the ability of WHI as an index of abdominal obesity to predict the risk of colon cancer in Japanese persons WHI is an efficient predictor (more effective than WC and BMI) of risk of colon cancer in females Katulanda et al34 Sri Lanka XUD Descriptive cross-sectional study; n = 4485; 1772:2713; ≥18 y XUD: distance between the lower border of xiphisternum and the center of umbilicus at the end of normal expiration; Cutoff: NR Relationship between obesity-associated risk factors for CVD and XUD, WC, BMI, and WHR XUD is significantly, but weakly, associated with obesity-associated risk factors for CVD, but the association was weaker than that of BMI, WC, and WHR with obesity-associated risk factors for CVD Kommuri et al35 USA C-index Longitudinal cohort study; n = 6814; 3213:3601; 45–84 y C-index = WC (m) / (0.109 × ✓ [Wt (kg) / Ht (m)]) Cutoff: NR To explore the association between various anthropometric measures and markers of subclinical atherosclerosis C-index was associated with subclinical atherosclerosis compared to other measures. However, the association was weaker than other anthropometric parameters Krakauer and Krakauer36 USA HI; ARI Cohort study; n = 16 034; 7696:8338; Mean age 43 y HI = HC of a given person normalized to a standard height and weight ARI = the sum of function values for each individual’s combination of anthropometric index scores, denoting natural logarithm of the combined estimated hazard from 4 independent indices: HC, BMI, ABSI, and HI Cutoff: NR To evaluate HI and ARI as predictors of mortality risk HI and ARI were consistent predictors of mortality hazard Li et al37 USA WTR Descriptive cross- sectional study; n = 6277; 2994:3283; ≥20 y WTR = WC (cm) / TC (cm) (TC = circumference of the mid-thigh, perpendicular to the long axis of the thigh) Cutoff: NR To examine whether WTR performed better than WHtR, WHR, WC, and BMIin relation to T2DM among US adults For the association with T2DM, WTR performed better than the other 4 indices in men, and WTR performed similarly to WHtR, WHR, and WC, but better than BMI, in women Motamed et al38 Iran AVI Descriptive cross- sectional study; n = 3201; 1826:1375; 40–79 y AVI= (2 × WC2 + 0.7 [WC – HC2]) / 1000 Cutoffs: NR To determine the performance of WC, WHR, WHtR, C-index, and AVI to predict risk of 10-y CVD events AVI had the lowest discriminatory accuracy to predict 10-y CV events, while C-index had the highest accuracy Ononamadu et al39 Nigeria PI Descriptive cross-sectional study; n=912; 436:476; 17–79 y PI = Wt (kg) / (Ht [cm])3 Cutoffs: hypertension and pre-hypertension: PI = 14.45, 13.69 (female); 16.38, 17.65 (male) To compare anthropometric indices of obesity as correlates and potential predictors of risk of hypertension and pre-hypertension BMI, WHtR, WC, and PI were the best predictors of hypertension risk, and BMI, WC, and PI of pre-hypertension risk Rahman and Adjeroh40 USA SBSI Longitudinal cohort study; n = 11 808; Sex: NR; 18–85 y SBSI = (Ht7/4 × WC5/6) / (BSA × VTC) Cutoff: NR To evaluate SBSI as a predictor of all-cause mortality SBSI is generally linear with age and increases with increasing mortality, when compared with other popular anthropometric indices of body shape Song et al12 Finland WHHR Cohort study; n = 46 651; 24 686:21 965; 24–99 y WHHR = WC (m) / (HC [m] × Ht [m]) Cutoffs: NR To compare the ability of anthropometric indicators of obesity to predict CVD mortality WHHtR was a stronger predictor of CVD mortality than BMI Vogel et al41 Brazil AFA Cohort study; n = 116; 72:44; 30–85 y AFA = (AC * TSF/2) – (3.14*[TSF]2/4) Cutoffs: 90th percentile (obesity) To evaluate the association between obesity indices and mortality among patients with ischemic heart failure AFA did not predict survival among heart failure patients Zhou et al42 China NC Descriptive cross-sectional study; n = 4201; 2508:1693; 20–85 y NC = measured with head erect and eyes facing forward, horizontally at the upper laryngeal bulge Cutoffs: males ≥37 cm, females ≥33 cm To investigate whether NC independently contributes to the prediction of cardiometabolic risks over BMI, WC, and WHR in Chinese individuals NC independently contributed to the prediction of cardiometabolic risks in Chinese individuals Author [Ref]; Country . Indices . Study population (study design; sample size;male:female ratio; age group) . Measurement (method; cutoff) . Comparison/objective . Findings . Abulmeaty et al24 Saudi Arabia HtWrR Descriptive cross-sectional; n = 390; 167:223; 18–50 y HtWrR = Ht (cm) / WC (cm) Cutoff: NR To investigate the prediction of long-term cardiometabolic risk using anthropometric parameters HtWrRdid not discriminate the risk in either sex Al-Gindan et al25 USA MCC; MTC Case-control cases: n = 416 (194: 222) 18–88 y Controls: n = 204 (94:110) 18–86 y MCC – maximum girth of calf MTC – measured at the midpoint between the inguinal crease and the proximal border of the patella Cutoff: NR To validate new anthropometric equations to estimate MRI-measured TATM/ TATFM/total body fat and compare these with existing prediction equations using older methods New anthropometric measurement–based equations correlated better with MRI-measured TATM and TATFM than existing equations Carlsson et al26 Sweden SAD; SADHR Longitudinal cohort study; n = 3741; 1751:1990; 60 y SAD = distance between table and the top of the body at the level of the iliac crest SADHR= SAD (cm) / Ht (cm) Cutoff: NR To compare anthropometric measures (BMI, WHR, WC, SAD, WHHtR, WHtR, and SADHR ) in their ability to predict ischemic CVD BMI, WC, and WHtR were weaker predictors than SAD and SADHR in predicting ischemic CVD Chagas et al27 Brazil AbCHR Descriptive cross-sectional study; n = 337; 213:124; 23–89 y AbCHR = AbC (cm) / HC (cm) AbC measured 1 cm above the iliac crest Cutoff: NR To evaluate the association of different anthropometric parameters with the coronary atherosclerotic burden AbCHR was not an independent risk factor for coronary atherosclerotic burden Chang et al28 China BRI Descriptive cross-sectional study; n = 11 345; 5253:6092; ≥ 35 y BRI = 364.2–365.5 × (1– [WC/2π]2 / [0.5 × Ht]2) Cutoff: 0.66 (male) and 0.67 (female) To assess the capacity of BRI to identify subjects with T2DM and determine whether it is superior to BMI, WC, and WHtR BRI was not superior to BMI, WC, or WHtR for predicting the presence of T2DM Chung et al29 Korea ABSI; LBSIZ Cohort study; n = 54 893; 23 769: 31 124; 20–103 y ABSI = WC (m) / (BMI2/3 × Ht1/2) LBSI = log(ABSI) LBSIZ = (LBSI–LBSI[mean])/LBSI (SD) Cutoff: ABSI = 0.68; LBSIZ = 1.32 (male) and 1.86 (female) To assess sensitivity of ABSI and LBSIZ based on BF% and its ability to predict hypertension and impaired health-related quality of life LBSIZ is a measure of abdominal obesity that can predict hypertension and impaired health-related quality of life, and was more generalizable than ABSI Fu et al30 China BAI Descriptive cross-sectional study; n = 4868; 2323:2545; 18–96 y BAI = (HC [cm] / Ht [cm]1.5) − 18 Cutoff: (for ≥1 cardiometabolic abnormality): BAI = 26.39 (male) and 31.29 (female) To correlate new anthropometric indices and cardiometabolic abnormalities BAI was significantly associated with cardiometabolic abnormalities Haghighatdoost et al31 Iran CUN-BAE Descriptive cross-sectional study; n = 9555; 4777:4778; Mean age 38.7±15.5 y CUN-BAE formula for BF% = –44.988 + (0.503 × age) + (10.689 × sex) + (3.172 × BMI) − (0.026 × BMI2) + (0.181 × BMI × sex) − (0.02 × BMI × age) – (0.005 × BMI2 × sex) + (0.00021 × BMI2 × age) Cutoff: NR To compare the predictive power of ABSI, BMI, WHtR, and CUN-BAE for MetS and cardiovascular disease risks CUN-BAE was a good predictor of body fat percentage; however, it was not the best anthropometric predictor of cardiovascular risks Jung et al32 Korea TC Descriptive cross-sectional study; n = 315 628; 199 423:116 205 30–79 y TC: measured immediately below the gluteal fold in left thigh Cutoff: NR To study the association between TC and T2DM, in relation to age and BMI Small TC was associated with T2DM, and this association was stronger among participants with a BMI <25 kg/m2. TC is a possible T2DM marker in lean populations Kaneko et al33 Japan WHI Descriptive cross-sectional study; n = 695; 433:262; 45–74 y WHI = WC (cm) / (Ht [m] × Ht [m]) Cutoff: ≥ 35 To determine the ability of WHI as an index of abdominal obesity to predict the risk of colon cancer in Japanese persons WHI is an efficient predictor (more effective than WC and BMI) of risk of colon cancer in females Katulanda et al34 Sri Lanka XUD Descriptive cross-sectional study; n = 4485; 1772:2713; ≥18 y XUD: distance between the lower border of xiphisternum and the center of umbilicus at the end of normal expiration; Cutoff: NR Relationship between obesity-associated risk factors for CVD and XUD, WC, BMI, and WHR XUD is significantly, but weakly, associated with obesity-associated risk factors for CVD, but the association was weaker than that of BMI, WC, and WHR with obesity-associated risk factors for CVD Kommuri et al35 USA C-index Longitudinal cohort study; n = 6814; 3213:3601; 45–84 y C-index = WC (m) / (0.109 × ✓ [Wt (kg) / Ht (m)]) Cutoff: NR To explore the association between various anthropometric measures and markers of subclinical atherosclerosis C-index was associated with subclinical atherosclerosis compared to other measures. However, the association was weaker than other anthropometric parameters Krakauer and Krakauer36 USA HI; ARI Cohort study; n = 16 034; 7696:8338; Mean age 43 y HI = HC of a given person normalized to a standard height and weight ARI = the sum of function values for each individual’s combination of anthropometric index scores, denoting natural logarithm of the combined estimated hazard from 4 independent indices: HC, BMI, ABSI, and HI Cutoff: NR To evaluate HI and ARI as predictors of mortality risk HI and ARI were consistent predictors of mortality hazard Li et al37 USA WTR Descriptive cross- sectional study; n = 6277; 2994:3283; ≥20 y WTR = WC (cm) / TC (cm) (TC = circumference of the mid-thigh, perpendicular to the long axis of the thigh) Cutoff: NR To examine whether WTR performed better than WHtR, WHR, WC, and BMIin relation to T2DM among US adults For the association with T2DM, WTR performed better than the other 4 indices in men, and WTR performed similarly to WHtR, WHR, and WC, but better than BMI, in women Motamed et al38 Iran AVI Descriptive cross- sectional study; n = 3201; 1826:1375; 40–79 y AVI= (2 × WC2 + 0.7 [WC – HC2]) / 1000 Cutoffs: NR To determine the performance of WC, WHR, WHtR, C-index, and AVI to predict risk of 10-y CVD events AVI had the lowest discriminatory accuracy to predict 10-y CV events, while C-index had the highest accuracy Ononamadu et al39 Nigeria PI Descriptive cross-sectional study; n=912; 436:476; 17–79 y PI = Wt (kg) / (Ht [cm])3 Cutoffs: hypertension and pre-hypertension: PI = 14.45, 13.69 (female); 16.38, 17.65 (male) To compare anthropometric indices of obesity as correlates and potential predictors of risk of hypertension and pre-hypertension BMI, WHtR, WC, and PI were the best predictors of hypertension risk, and BMI, WC, and PI of pre-hypertension risk Rahman and Adjeroh40 USA SBSI Longitudinal cohort study; n = 11 808; Sex: NR; 18–85 y SBSI = (Ht7/4 × WC5/6) / (BSA × VTC) Cutoff: NR To evaluate SBSI as a predictor of all-cause mortality SBSI is generally linear with age and increases with increasing mortality, when compared with other popular anthropometric indices of body shape Song et al12 Finland WHHR Cohort study; n = 46 651; 24 686:21 965; 24–99 y WHHR = WC (m) / (HC [m] × Ht [m]) Cutoffs: NR To compare the ability of anthropometric indicators of obesity to predict CVD mortality WHHtR was a stronger predictor of CVD mortality than BMI Vogel et al41 Brazil AFA Cohort study; n = 116; 72:44; 30–85 y AFA = (AC * TSF/2) – (3.14*[TSF]2/4) Cutoffs: 90th percentile (obesity) To evaluate the association between obesity indices and mortality among patients with ischemic heart failure AFA did not predict survival among heart failure patients Zhou et al42 China NC Descriptive cross-sectional study; n = 4201; 2508:1693; 20–85 y NC = measured with head erect and eyes facing forward, horizontally at the upper laryngeal bulge Cutoffs: males ≥37 cm, females ≥33 cm To investigate whether NC independently contributes to the prediction of cardiometabolic risks over BMI, WC, and WHR in Chinese individuals NC independently contributed to the prediction of cardiometabolic risks in Chinese individuals AbC, abdominal circumference, AbCHR, abdominal circumference to hip ratio; ABSI, a body shape index; ABSIz, a body shape index z-score AC, arm circumference; AFA, arm fat area; AH, abdominal height; ARI, anthropometric risk index; AUC, area under the curve; AVI, abdominal volume index; BAI, body adiposity index; BF, body fat; BMI, body mass index; BRI, body roundness index; BSA, body surface area; C-index, conicity index; CUN-BAE, Clinica Universidad de Navarra-Body Adiposity Estimator; CV, cardiovascular; CVD, cardiovascular disease; F, female; HC, hip circumference; HCNC, height-corrected neck circumference; HI, hip index; Ht, height; HtWrR, height to wrist ratio; LBSIZ, log-transformed body shape index; M, male; MAC, mid-arm circumference; MAMA, mid-arm muscular area; MCC, mid-calf circumference; MetS, metabolic syndrome; MRI, magnetic resonance imaging; MTC, mid-thigh circumference; NC, neck circumference; NR, not reported; OSA, obstructive sleep apnea; PCOS, polycystic ovarian syndrome; PI, ponderal index; SAD, sagittal abdominal diameter; SADHR, sagittal abdominal diameter to height ratio; SBSI, surface-based body shape index; TATFM, total adipose tissue fat mass; TATM, total adipose tissue mass; TC, thigh circumference; T2DM, type 2 diabetes mellitus; TSF, triceps skinfold thickness; VAT, visceral adipose tissue; VTC, vertical trunk circumference; WC, waist circumference; WHHR, waist circumference to hip circumference to height ratio; WHI, waist to height index; WHR, waist to hip ratio; WHtR, waist to height ratio; Wt, weight; WTR, waist circumference to thigh circumference ratio; XUD, xiphisternum to umbilicus distance. Open in new tab Furthermore, when a single study reported several anthropometric indices, only details pertaining to the novel anthropometric parameters described in the study were included. In addition, where possible the first article that proposed a particular index was also identified via retrospective search of citation, with differences in the original concept and present index being highlighted. RESULTS Literature search and study characteristics The literature search yielded 2472 articles, and the title and abstract of these papers were screened for relevance. Full-text copies were obtained for 96 articles and after reading, 66 studies were deemed eligible to be included in the final analysis. A summary of the search strategy is presented in Figure 1. Data on novel anthropometric parameters were derived from 26 countries (Australia, Brazil, Bulgaria, China, Colombia, Denmark, Finland, India, Iran, Italy, Japan, Malaysia, Netherland, Nigeria, Norway, Republic of Korea, Mexico, Portugal, Saudi Arabia, Singapore, Spain, Sri Lanka, Sweden, Turkey, UK, and USA). The majority were descriptive cross-sectional studies (n = 43, 65.2%), while 22 (33.3%) were cohort studies. Age range of the study populations was 17–103 years, while sample size varied from 45 to 384 612. The literature search identified 25 novel anthropometric parameters, all of which are listed in Figure 2. For each of the novel anthropometric parameters, the study that described the parameter in the largest cohort of participants was included in Table 1,12,24–42 with a description of the relevant study and the anthropometric parameter. Each parameter is described in detail in the following section. Figure 2 Open in new tabDownload slide List of the novel anthropometric parameters identified in this review. Figure 2 Open in new tabDownload slide List of the novel anthropometric parameters identified in this review. Indices based on a single anthropometric measurement Mid-calf circumference and mid-thigh circumference The mid-calf circumference is defined as the maximum girth of the calf25 and correlates strongly with magnetic resonance imaging (MRI)-measured total adipose tissue mass and total adipose tissue fat mass.25 Mid-thigh circumference was measured at the midpoint between the inguinal crease and the proximal border of the patella,25 and correlated strongly with MRI-measured total adipose tissue mass and total adipose tissue fat mass.25 However, it is important to acknowledge that although mid-calf circumference is defined as the maximal girth of the calf, the point of measurement may not be the mid-calf. Neck circumference In a study conducted by Assyov et al,43 neck circumference (NC) was measured between the mid-cervical spine and mid-anterior neck just below the laryngeal prominence among adults with severe obesity (BMI > 30 kg/m2). They concluded that NC was more effective than WC at distinguishing T2DM (area under the curve [AUC] = 0.758), insulin resistance (AUC = 0.757), metabolic syndrome (Met S; AUC = 0.724), and hypertension (AUC = 0.763) in those with severe obesity. NC was used by Akin et al44 to investigate the relationship between overactive bladder in women with MetS, with a high NC associated with an overactive bladder (AUC = 0.73). A large NC (measured just above the cricoid cartilage and perpendicular to the long axis of the neck) has also been associated with cardiovascular risk factors (odds ratio [OR]: 1.1–2.6) in older people45 and with metabolic factors such as pre-diabetes (OR: 1.18–1.26).46 According to Ozkaya and Tunckale,47 there was a significant correlation between NC and obesity (r = 0.24–0.68), as defined by traditional anthropometric parameters. NC has also shown positive correlations with WC, BMI, and MetS in Chinese individuals with T2DM.48 The authors defined NC cutoffs for overweight (males ≥38 cm, females ≥35 cm) and MetS (males ≥39 cm, females ≥35 cm). In another Chinese study, NC was significantly associated with cardiometabolic risk factors and independently contributed to the prediction of cardiometabolic risks (OR: 1.29–1.44) beyond the classical anthropometric indices.42 However, Chagas et al27 concluded that NC in patients undergoing coronary angiography for suspected coronary artery disease was not an independent risk factor for atherosclerotic burden after multivariate analysis (r = 0.09). It is important to note that authors have used different sites in the assessment of NC, which may have influenced the final results. Sagittal abdominal diameter The sagittal abdominal diameter (SAD) has been measured at different anatomical sites in various studies including umbilical level, highest abdominal diameter, the narrowest point between the last rib and iliac crest, and the midpoint between the iliac crests.49 SAD measured at the midpoint between the iliac crests strongly correlated with cardiometabolic risk factors in elderly men (r = 0.107–0.480).49 Dahlen et al50 prospectively explored how SAD predicted subclinical organ damage in patients with T2DM, with SAD measured with the patients in the supine position with bent knees, at the highest point of the abdomen. Results showed that SAD predicted arterial stiffness over 4 years in patients with T2DM (r = 0.184). In a subsequent study, Dahlen et al51 found that SAD was a good predictor of inflammation (r = 0.29–0.31) and subclinical organ damage (r = 0.11–0.21) in middle-aged patients with T2DM. Other studies have shown that SAD is a good predictor of central obesity among women (r = 0.79),52 but there were no significant correlations between body composition as measured by SAD and lung function.53 SAD ≥25 cm was a significant and independent risk factor (hazard ratio [HR]: 2.81) that predicted major cardiovascular events in patients with T2DM compared with WC (HR: 1.44).54 These findings have also been verified in a cohort study involving adults older than 60 years followed up for 11 years.55 An inverse relationship was found between short sleep duration and SAD among Swedish females (−0.46 cm/h),56 and SAD accurately estimated accumulation of epicardial adipose tissue and cardiovascular risk in a study conducted among premenopausal Brazilian females (AUC: 0.81).57 Thigh circumference A small thigh circumference (TC) has been associated with an increased risk of developing heart disease (TC < 56 cm in males and TC < 68 cm in women) or premature death (TC < 62 cm) in a prospective cohort study among males and females aged 35–65 years, followed up for 10.0–12.5 years.58 A similar study among 2484 participants aged 50–75 years concluded that a larger TC was associated with a lower risk of T2DM in women, independent of BMI, age, and WC.59 A large population study (n = 199 243) by Jung et al32 also confirmed that a small TC was associated with T2DM (AUC: 0.795). The study further concluded that this association was stronger among participants with a BMI less than 25 kg/m2, making TC a possible useful T2DM marker in lean populations. In all studies, the TC was measured immediately below the gluteal fold of the left leg, in contrast to the mid-thigh circumference measurement in previously discussed studies. Xiphisternum to umbilicus distance The xiphisternum to umbilicus distance (XUD) was first described by Katulanda et al,34 who investigated the relationship between XUD and cardiovascular disease (CVD). XUD was defined as the distance between the lower border of the xiphisternum and the center of umbilicus at the end of normal expiration. The study concluded that XUD was significantly, but weakly (AUC for ≥2 cardiovascular risk factors: 0.62), associated with obesity-associated risk factors for CVD (AUC less than for BMI, WC, WHR). The authors also observed a significant correlation between XUD and BMI, WC, and WHR (P < 0.001). Indices adjusted for height Ponderal index The ponderal index was originally described by Rohrer as a measure of intrauterine growth retardation in infants.60 It is derived by dividing weight in kilograms by (height)3 in centimeters (Table 1).12,24–42 Ononamadu et al39 compared ponderal index and other indices as predictors of risk of hypertension and pre-hypertension in a cross-sectional study in Nigeria. PI (AUC: 0.52–0.68), together with BMI (AUC: 0.52–0.73) and WC (AUC: 0.51–0.61), were the best predictors of hypertension and pre-hypertension risk; however, a combination of indices in a regression model did not improve their performance as predictors. Sagittal abdominal diameter to height ratio The sagittal abdominal diameter to height ratio (SADHR) was used to predict ischemic CVD risk in an 11-year longitudinal cohort study comprising 3471 Swedish people26 with SAD measured as the perpendicular distance between the table and top of the body at the level of the iliac crest in supine position, measured after normal expiration, using a ruler and spirit level. SADHR was a strong predictor of ischemic cardiovascular disease risk in the cohort. BMI (HR: 0.99–1.08), WC (HR: 0.99–1.07), and WHtR (HR: 1.04–1.12) were weaker predictors than SAD (HR: 1.05–1.16) and SADHR (HR: 1.10–1.19) in predicting ischemic CVD. Waist circumference to hip circumference to height ratio Waist circumference to hip circumference to height ratio (WHHR) was first described by Rosenblad et al61 and is calculated as the ratio between WHR and height (Table 1).12,24–42,61 A descriptive cross-sectional study involving 4868 Chinese adults showed a significant association between WHHR with cardiometabolic abnormalities (hypertension [AUC: 0.67–0.69], T2DM [AUC: 0.67], and dyslipidemia [AUC: 0.58–0.64]); however, other anthropometric parameters (BMI, WC, WHR, WHtR) showed a better association.30 The authors recommended a cutoff WHHR value of 0.51 in males and 0.53 in females for the presence of 1 or more cardiometabolic abnormalities. Song et al12 compared the ability of WHHR and other anthropometric indices to predict CVD mortality in a longitudinal cohort study (7.9 y) involving a population of 50 093 adults from 12 prospective studies conducted in 4 different European countries (Finland, Sweden, Turkey, and UK). This large population study revealed that WC (HR: 1.29–1.49), WHR (HR: 1.28–1.45), WHtR (HR: 1.35–1.52), and WHHR (HR: 1.37–1.45) were stronger predictors for CVD mortality than a body shape index (ABSI) (HR: 1.32–1.34) or BMI (HR: 1.19–1.37). Similar observations were reported in a study by Carlsson et al.26,55 In a cohort of 3741 adults without CVD followed up for 11 years, WHHR (HR: 1.20) and WHR (HR: 1.14) were the best predictors of CVD in normal-weight women, and among overweight/obese individuals. WHHR was the strongest predictor after adjustments for CVD risk factors in men. A study conducted in a cohort of US adults showed that WHHR had a lower association and was an inferior discriminator of incident T2DM (HR: 1.26–1.61) among all race-sex groups, compared with BMI (HR: 1.56–1.76), WC (HR: 1.56–1.88), WHtR (HR: 1.57–1.86), and WHR (HR: 1.26–1.77).62 All the above authors used the same measures to derive WHHR. Waist to height index Waist to height index is calculated using the WC divided by height squared (Table 1),12,24–42 and was first described by Kaneko et al33 in a cohort of Japanese patients who underwent colonoscopy. Waist to height index was an efficient predictor (OR: 1.32; P < 0.05) (more than WC and BMI) of risk of colonic cancer among the female patients.33 The study recommended that women aged ≥55 years and/or with waist to height index ≥35 should undergo elective colonic endoscopy. All the above authors used the same calculation to derive waist to height index. Indices adjusted for height and weight Conicity index Conicity index (C-index) is calculated according to WC, weight, and height (Table 1).12,24–42 It was first described by Valdez63 in 1991. This measurement has been used to investigate the prediction of long-term cardiometabolic risk among middle-aged males and females, and demonstrated good clinical discriminatory value for long-term cardiometabolic risk with an AUC of 0.817.24 Another study evaluated the performance of C-index in discriminating high coronary risk in women64 and showed that C-index was the indicator with highest discriminatory power in this cohort of women (AUC: 0.76). According to Chakraborty and Bose,65 C-index showed no advantage over other adiposity measures in the prediction of hypertension among slum-dwelling Bengalee men in Kolkata. Chakraborty used 1.25 as the cutoff value for C-index. Similar findings were observed by Ononamadu et al,39 who compared different anthropometric indices of obesity as correlates and potential predictors of risk of hypertension and pre-hypertension in a cross-sectional study. However, a high C-index has been found to be associated with a high risk of hypertension (OR: 4.3) among pre-university students.66 C-index has also been used to discriminate MetS in Brazilian women with polycystic ovarian syndrome.67 The authors concluded that WC (AUC: 0.83) and WHtR (AUC: 0.82) were more effective than C-index (AUC: 0.74) at predicting MetS, with similar findings observed in a study among Chinese adults.68 Kommuri et al35 explored the associations between various anthropometric measures and markers of subclinical atherosclerosis in a longitudinal cohort study and concluded that C-index was a less consistent marker – in its association with various markers of subclinical atherosclerosis – than other anthropometric measures. However, another study showed that C-index (AUC: 0.67– 0.76) had the highest discriminatory accuracy to predict 10-year cardiovascular events compared with WC (AUC: 0.57–0.59), WHtR (AUC: 0.62–0.65), and abdominal volume index (AVI; AUC, 0.57–0.59).38 C-index was a useful marker of inflammatory status, abdominal fat mass, and protein energy wasting in post-hemodialysis patients.69 All the above authors used the same calculation to derive C-index. Hip index Hip index is defined as the hip circumference (HC) of a given person normalized to a standard height and weight (Table 1).12,24–42 In analyses of data from the longitudinal cohort studies US National Health and Nutrition Examination Survey (NHANES) III and Atherosclerosis Risk in Communities Study (ARIC), involving 16 034 adults, Hip index was a consistent predictor of mortality hazard (“U”-shaped relationship); however, in both cohorts, BMI (HR: 1.06–1.11) and ABSI (HR: 1.16–1.26) were better nonlinear indicators of mortality hazard.36 Furthermore, since hip index is calculated according to the standard height and weight of a population, it cannot be determined for populations from countries where national data are not available. Indices to estimate percentage body fat Body adiposity index Body adiposity index (BAI) is an anthropometric parameter derived from HC and height (Table 1).12,24–42 It was first described by Bergman et al70 as a direct estimate of percentage adiposity. Belarmino et al71 evaluated the performance of BAI in estimating body fat percentage (BF%) (measured by air displacement plethysmography) in severely obese Brazilian patients. BAI did not provide an accurate estimate of BF% in this study (level of agreement: 5.7%–16.0%) and has been reported to poorly predict body fat indices in obese women72 and in athletes.73 BAI (OR: 2.3) showed significant, but considerably lower, correlation than other adiposity measures (WHtR [OR: 4.0], BMI [OR: 3.3], WHR [OR: 4.2]) in the prediction of hypertension among slum-dwelling Bengalee men in Kolkata.65 However, BMI (65%) and BAI (45%) were significant predictors of hypertension and subclinical organ damage in adult men in a prospective cohort study after 8 years of follow-up.74 However, in a cross-sectional study, Ononamadu et al39 showed that BAI correlates poorly with blood pressure (r = 0.12–0.18). Fu et al30 found a significant correlation between BAI and cardiometabolic abnormalities (hypertension, T2DM, and dyslipidemia) in a descriptive cross-sectional study involving 4868 subjects. However, the association was weaker than traditional anthropometric parameters.30,75 Similarly, BAI was not superior to traditional obesity indices for predicting MetS.68,76 A study conducted by Garcia et al77 found that BAI was a statistically significant predictor of cardiovascular risk (OR: 1.6–9.3) (criteria described by the National Cholesterol Education Program). In a prospective follow-up study of 7 years’ duration, BAI was significantly associated with the presence of T2DM in men (OR: 1.9), but not in women (OR: 1.0).78 However, BAI had lower associations and was an inferior discriminator of incident T2DM.62 Furthermore, BAI was not a significant predictor of survival among ischemic heart failure patients.41 All the above authors used the same calculation to derive BAI. Body roundness index Body roundness index (BRI), first described by Thomas et al,79 is calculated according to the WC and height (Table 1)12,24–42 and has been shown to slightly improve predictions of BF% and the percentage of visceral adipose tissue, compared with the traditional metrics of BMI, WC, or HC.79 Chang et al28 used BRI to identify participants with T2DM, and compared this index with traditional anthropometric indices. Although BRI (OR: 1.8–1.9) showed potential for use as an alternative obesity measure in the assessment of T2DM, it was not superior to BMI (OR: 1.6), WC (OR: 1.8–1.9), or WHtR (OR: 2.4–2.7) for predicting T2DM, in a rural Chinese population. BRI has also shown predictive value in MetS, especially among males.68 However, the index was not superior to traditional obesity indices for predicting MetS.76 Santos et al73 compared BF% (bioelectrical impedance) with novel anthropometric measurements, including BRI, and found that these indices were poor predictors of BF% in athletes. All the above authors used the same calculation to derive BRI. Clinica Universidad de Navarra-body adiposity estimator The Clinica Universidad de Navarra-body adiposity estimator (CUN-BAE) formula is used to estimate BF% and is based on age, sex (where male = 0 and female = 1), and BMI (Table 1).12,24–42,80 CUN-BAE which was first described by Gomez-Ambrosi et al,81 has shown a very good correlation (r = 0.89, P < 0.001) with BF% (measured by air displacement plethysmography). A study conducted in Iran showed that CUN-BAE was a predictor for CVD risks and MetS (OR: 0.9–1.2); however it was not the best predictor of CVD risk in the Iranian population, compared with other traditional anthropometric parameters.31 However, the Hordaland Health Study,82 a prospective 6-year follow-up study in Norway, identified that CUN-BAE is more strongly associated with future risk of T2DM (4.3–5.4) and CVD (OR: 1.9–2.1) than with BMI (OR: 1.2–2.1) in an analysis stratified according to sex. A study conducted in Spain, to evaluate the relationship between CUN-BAE formula in comparison with BMI in the prediction of T2DM and hypertension, showed that the overall correlation between BMI and CUN-BAE was not good (R2 = 0.48), but improved when age and gender were taken into account (R2 > 0.90).80 This study also concluded that CUN-BAE was a better predictor than BMI for hypertension and T2DM.80 All the above authors used the same formula to calculate CUN-BAE. Other indices Abdominal circumference to hip ratio Abdominal circumference to hip ratio (AbCHR) is defined as the ratio between the abdominal circumference and hip circumference. Chagas et al,27 who first described the AbCHR, evaluated the association between AbCHR (and other standard anthropometric measurements) and coronary atherosclerosis burden (coronary angiography Friesinger score). In this study, abdominal circumference was measured 1 cm above the iliac crest and the maximum circumference between hips and buttocks was considered to be the hip circumference. In the above study, AbCHR was not an independent risk factor for coronary atherosclerotic burden (r = 0.102, P = 0.061). All the above authors used the same calculation to derive AbCHR. Arm fat area Arm fat area is calculated according to arm circumference and triceps skin fold thickness (Table 1).12,24–42 According to Vogel et al,41 arm fat area did not predict survival among ischemic heart failure patients. In this study, arm circumference and triceps skin fold thickness was measured at the midpoint between the acromion and the olecranon, with the arm extended down the side of the body and the palm of the hand facing the thigh. The above study utilized the same formula originally proposed by Gurney and Jelliffe83 to calculate the arm fat area. ABSI, ABSI z-score, and log-transformed ABSI z-score ABSI is a recently introduced marker of abdominal adiposity derived from WC, BMI, and height (Table 1).12,24–42 ABSI measures WC in relation to weight and height and thus can be a measure of abdominal obesity independent of weight, height, or BMI.29 Studies have shown that ABSI could be used to define sarcopenia in overweight/obese individuals, where those with a lower ABSI (r = −0.37) have a significantly greater fat-free mass index.84 However, ABSI poorly predicts body fat percentage in athletes (R2 = 0.22).73 According to Chang et al,28 ABSI (OR: 1.51–1.55) was not superior to traditional anthropometric measurements (BMI [OR: 1.57], WC [OR: 1.79–1.90], and WHtR [OR: 2.40–2.67]) in predicting the presence of T2DM in a large cohort of Chinese adults (n = 11 345). The same conclusion was reached in another study conducted among adults from the USA.62 Furthermore, Fu et al30 found no significant correlation between ABSI with cardiometabolic abnormalities (hypertension [OR: 0.07–0.08], T2DM [OR: 0.08], and dyslipidemia [OR: 0.07–0.08]) in a descriptive cross-sectional study involving 4868 participants. Another study including 9555 individuals from Iran also concluded that ABSI was a weak predictor of cardiovascular disease risks (hypertension [OR: 0.9–1.1], hyperglycemia [OR: 0.8–1.1], hypercholesterolemia [OR: 0.8–1.2]) and MetS (OR: 1.7–1.9).31 These findings were confirmed by a study among Chinese adults, where ABSI did not show a predictive value for MetS for either sex.68 However, a recent study among 6081 Caucasian adults, which tested the separate and joint contribution of ABSI and BMI to high triglyceride levels, low levels of high-density lipoprotein cholesterol, high blood pressure, and high fasting glucose and visceral abdominal fat thickness (by ultrasound), found that ABSI was independently associated with all outcomes.85 ABSI has been shown to be a weaker predictor of CVD mortality (HR: 1.32–1.34) than WC (HR: 1.29–1.49), WHR (HR: 1.28–1.45), WHtR (HR: 1.35–1.52), and WHHR (HR: 1.34–1.45).12 However, US studies showed that all-cause mortality increases with increasing ABSI among US citizens.36,40 These findings were also confirmed among a cohort of Australian adults.86 The formula used by Krakauer and Krakauer,87 who first proposed ABSI, was used in all of the above studies to derive ABSI. To improve the predictive ability of ABSI, several authors have proposed the use of ABSI z-score and log-transformed ABSI z-score. ABSI z-score has also been used to evaluate all-cause mortality and found to be a consistent predictor of mortality hazard (HR: 1.13–1.18) compared to other measures of abdominal obesity such as BMI (HR: 1.00), WC (HR: 1.09), and WHtR (HR: 1.11).88 Log-transformed ABSI z-score has been shown to be an independent predictor of hypertension (OR: 1.17–1.22), impaired health-related quality of life (OR: 1.11–1.27), and obesity (OR: 1.32–1.86).29 Anthropometric risk index Anthropometric risk index, first defined by Krakauer and Krakauer, is the sum of function values for each individual’s combination of anthropometric index scores, denoting the natural logarithm of the combined estimated hazard from the 4 independent indices HC, BMI, ABSI, and hip index (Table 1).12,24–42 In analyses of data from the longitudinal cohort studies NHANES III and ARIC, involving 16 034 adults, anthropometric risk index was a consistent predictor of mortality hazard (HR: 1.43–1.46) and a substantially better predictor of mortality risk than any of the individual anthropometric indices tested.36 Abdominal volume index Abdominal volume index (AVI) was first described by Guerrero-Romero and Rodríguez-Morán,89 for the purpose of determining the obesity-associated risk of diabetes, where AVI is derived from WC and HC (Table 1).12,24–42 The authors concluded that AVI is strongly related to impaired glucose tolerance (OR: 1.6) and T2DM (OR: 2.1),89 with similar results observed in a cohort of Mexican women.90 The same measurement was used by Abulmeaty et al24 to investigate the prediction of long-term cardiometabolic risk (calculated using 5 different CVD risk scoring systems). The results showed that AVI was significantly positively correlated with long-term CVD risk scores in both men (r = 0.229) and women (r = 0.345). In another study, AVI (AUC: 0.58) was not as effective as the C-index (AUC: 0.67–0.739) as a predictor of 10-year cardiovascular events.38 Another prospective cohort study, where the participants were followed up for 4.5 years for the development of MetS, both BMI (AUC: 0.73–0.75) and AVI (AUC: 0.79) were superior to the other anthropometric indices for predicting MetS in both men and women.68 All the above-stated authors used the same calculation to derive AVI. Height to wrist ratio Height to wrist ratio is the ratio between height and wrist circumference (Table 1).12,24–42 This measurement has been used to investigate the prediction of long-term cardiometabolic risk calculated using 5 different CVD risk scoring systems.24 The study concluded that height to wrist ratio did not discriminate the risk in either sex, for any of the CVD risk scoring systems evaluated. Surface-based body shape index Surface-based body shape index is derived from height, WC, body surface area, and vertical trunk circumference (Table 1).12,24–42 Rahman and Adjeroh,40 who first described the surface-based body shape index, evaluated it as a predictor of all-cause mortality. They measured vertical trunk circumference using a tape from the shoulder, through the crotch, and back to the shoulder while the participant stands fully erect with the weight distributed equally on both feet and the arms hanging freely downwards. Body surface area was calculated as a product of 0.00949 × weight (0.441) in kilograms × height (0.655) in meters. The study showed that the surface-based body shape index was generally linear with age, and increased with increasing mortality, and was more effective (HR: 2.3) than other popular anthropometric indices of body shape, including BMI (HR: 0.91), WC (HR: 1.3), and ABSI (HR: 2.3). Waist circumference to thigh circumference ratio Waist circumference to thigh circumference ratio (WTR) is defined as the ratio between WC and mid-TC (Table 1).12,24–42 Duarte-Rojo et al91 attempted to identify the predictive ability of WC, WHR, and WTR in relation to severe acute pancreatitis. Umbilical WC was the circumference at the level of the umbilicus, or above the iliac crests when displacement of the umbilicus was noticed. Findings suggested that WC, WHR, and WTR were all accurate predictors of severe acute pancreatitis. However, umbilical WC was the best predictor and the only variable retained in the multivariate analysis. In another study, WTR was associated (OR: 4.21–4.68) with the presence of peripheral vascular disease (defined as an ankle to brachial pressure index of <0.90 in at least one leg) in both males and females, and the association was much stronger than that with WC, especially in males (OR: 1.06–1.09).92 Several studies have evaluated the relationship between WTR and T2DM. Data analysis from the US NHANES III (1998–1994) showed that WTR (AUC: 0.83) performed better than 4 traditional indices in men (WHtR [AUC: 0.78], WHR [AUC: 0.79], WC [AUC: 0.76], and BMI [AUC: 0.72]). WTR (AUC: 0.80) performed similarly to WHtR (AUC: 0.80), WHR (AUC: 0.79), and WC (AUC: 0.78), but better than BMI (AUC: 0.73), in women for the association with T2DM.37 Similar results were observed in a study conducted among a cohort of 1055 patients from North India, where out of several anthropometric measurements (including WC, HC, WHR), WTR (r = 0.324–0.377, P < 0.01) correlated significantly and positively with blood glucose (fasting, random, and postprandial) (P < 0.001), suggesting that it is the best predictor of T2DM.93 The study recommended a WTR of 2.3 as a quick, noninvasive diagnostic tool for T2DM. WTR was originally proposed by Kahn et al,94 who measured WC at the midpoint between iliac crest and lower ribs, while mid-TC was measured midway between lateral inguinal fold and mid-patella. Variations were noted in the study by Duarte-Rojo et al,91 which used umbilical WC and upper TC to calculate WTR. DISCUSSION This is the first paper to systematically evaluate the literature in order to identify, compare, and contrast novel anthropometric approaches to assessing overweight and obesity in adults. Overweight and obesity are associated with an accumulation of body fat that may have a negative impact on health, with the latter generally defined as a body fat percentage of greater than 25% for men and greater than 32% for women. The 4-compartment analysis process is currently considered the “reference method” for body composition assessment95 and incorporates independent measurement of bone mineral content, total body water, and body density to formulate a body fat prediction. However, it is neither practical nor feasible as a clinical measurement in most healthcare settings, as the process is both costly and time consuming. Hence, anthropometric parameters such as BMI, WC, HC, WHR, and WHtR are more typically used to define obesity in adults.15–17,20,21 An ideal anthropometric measure for defining adiposity should be both simple and accurate.22 Several novel anthropometric parameters identified in the present analysis, such as the BRI, ABSI, CUN-BAE formula, and C-index, used complex calculations to derive level of adiposity.28,29,31,35 Furthermore, other anthropometric parameters relied upon several body measurements that were both complex and difficult to measure with precision. This lack of simplicity is likely to be a major barrier in the application of some of the novel anthropometric parameters cited, especially as a screening tool for overweight and obesity in primary and secondary healthcare settings. Accuracy in the measurement of adiposity is another important characteristic of an ideal anthropometric parameter. The standard used to compare the novel anthropometric parameter varied in the studies included in the present analysis, with only a few studies using body fat percentage as the comparator.25,31 As described earlier, the 4-compartment model is the ideal reference method for measuring body fat. However, most of the studies cited used single approaches such as Bioelectrical impedance analysis,73 air displacement plethysmography,71 and magnetic resonance imaging25 to determine body fat, and used it as the reference method. Furthermore, most studies have only evaluated the relationship of body fat with the presence of cardiovascular/metabolic diseases such as T2DM, dyslipidemia, hypertension, and CVD,28–30,32,34,37,39,42,55 and/or studied the relationship with cardiovascular/metabolic risks, by using scoring systems such as the Framingham risk scoring system and/or the American College of Cardiology and American Heart Association risk score.24,38 Prospective follow-up for the outcomes and/or mortality were only evaluated in a few of the studies.12,40,55,88 Similarly, few studies evaluated the relationship between the anthropometric measurement and arterial stiffness, inflammatory markers, and atherosclerotic burden, as assessed by angiography.27,35 Hence, there is a need to further evaluate these novel anthropometric parameters to better understand their ability to predict body fat/adiposity, ideally as defined by the “gold standard” 4-compartment model. There is also a need for further prospective follow-up studies to understand the relationship of these anthropometric parameters with metabolic risk, cardiovascular/metabolic outcomes, and mortality. The study populations included participants from 26 different countries, across 6 continents, with subjects from diverse sociodemographic and ethnic backgrounds. It is important to appreciate that obesity-associated adverse health outcomes vary among these different populations and ethnic groups. For example, south Asians have a higher cardiovascular disease risk than white Caucasians at a given BMI and WC value,96–98 and this has led to the development of different ethnic/population specific cutoff values for BMI. Hence, although some of the novel anthropometric parameters have shown a positive relationship between risk factors and disease outcomes, the derived cutoffs may not be applicable across all populations and ethnic groups. In addition, a positive or negative association identified with any given anthropometric parameter may not be applicable to all populations. This limits their usage until further evaluation is undertaken and ethnic/population specific cutoff values are derived through future studies. Furthermore, although total adiposity is considered a risk factor for cardiovascular and metabolic disease, many studies have shown that fat distribution influences metabolism independent of the effects of total body fat stores.99 The accumulation of fat in the abdominal area, particularly in the visceral fat compartment, seems to be associated with an increased risk of complications such as insulin resistance, T2DM, dyslipidemia, and atherosclerosis.99 Furthermore, ectopic fat deposits around internal organs, and pericardial and peri-aortic tissues, for example, have also been shown to be associated with cardiovascular disease, cancer, and mortality in prospective follow-up studies.100 Similarly, although abdominal visceral adipose tissue and abdominal subcutaneous adipose tissue are both associated with adverse cardiometabolic risk factors, the association is much stronger for visceral adipose tissue.101 Hence, an ideal adiposity measure should also reflect body fat distribution. Most of the novel anthropometric parameters failed to distinguish fat distribution from body composition. In addition to body fat percentage and fat distribution, recent studies have also identified the importance of muscle mass on cardiometabolic risk and mortality. A large population-based cohort study involving more than 11 000 adults in the USA showed that at a given level of BMI, those with low muscle mass had higher total body fat percentage and WC, were more likely to have T2DM, and had an increased risk of death.102 The reduced survival for people with normal BMI, compared with survival in overweight persons, is possibly explained by loss of muscle mass in the former.103 Few anthropometric parameters identified in the present review are also affected by muscle mass in addition to the fat content measured. For example, a low TC has been shown to be associated with an increased risk of developing heart disease or premature death, whereas a larger TC was associated with a lower risk of T2DM, independent of BMI, age, and WC.58,59 Even analyses using direct measures of adiposity likely underestimate the risks of excess body fat as they fail to account for the increases in muscle mass that typically accompanies obesity.102 These changes in muscle mass have an impact on mortality risk that is independent and in a direction opposite to the effect of increased body fat. Hence, muscle mass is also an important factor that needs to be considered in an anthropometric measure, in addition to identifying total adiposity and fat distribution. This systematic review has several notable strengths, including being the first to critically review novel anthropometric parameters of adiposity and present the available evidence for each parameter for different populations. This makes it possible for future researchers to select the ideal parameter, based on the purpose of the study and the population involved. Furthermore, the identified gaps in the present knowledge will serve to guide future researchers. A potential limitation of the present analyses is the fact that the search was limited to a single database (PubMed); nevertheless, a large number of studies and different anthropometric parameters were identified. Furthermore, several novel anthropometric variables could not be compared with a direct measure of body fat, owing to a lack of published research on such parameters. Hence, an estimation of disease risk was employed – ie, for each of these parameters, their predictive value was estimated, rather than their accuracy as an indirect estimate of body composition. CONCLUSION In conclusion, the novel anthropometric parameters defining obesity identified in the present study showed variable correlation with obesity and/or related metabolic risk factors. Some of the parameters involved complex calculations, while others were derived from traditional anthropometric measurements. Owing to the absence of studies comparing most novel anthropometric parameters with direct measurements of body fat, further research is required in order to determine their accuracy and precision. Acknowledgments Author contributions. R.J., P.R., T.R., and A.P.H. substantially contributed to the general idea and design of the study. R.J., P.R., T.R., and Y.M. took part in designing the protocol. R.J., P.R., Y.M., and S.W. planned the data analysis. P.R., Y.M., S.W., and A.P.H. drafted the manuscript. All authors have read and consented to the manuscript. Funding. The above work was not supported by any external funding. Declaration of interest. The authors have no relevant interests to declare. Supporting Information The following Supporting Information is available through the online version of this article at the publisher’s website. Table S1PRISMA checklist References 1 Obesity and overweight: Fact Sheet. Available at: http://www.who.int/mediacentre/factsheets/fs311/en/. Accessed January 3, 2018. 2 Katulanda P Jayawardena MA Sheriff MH , et al. . Prevalence of overweight and obesity in Sri Lankan adults . Obesity Rev. 2010 ; 11 : 751 – 756 . Google Scholar Crossref Search ADS WorldCat 3 Krauss R Winston M Fletcher B , et al. . Obesity: impact on cardiovascular disease . Circulation. 1998 ; 98 : 1472 – 1476 . Google Scholar Crossref Search ADS WorldCat 4 Lancet T. Curbing the obesity epidemic. Lancet. 2006;367:1549. 5 Pi‐Sunyer FX. The obesity epidemic: pathophysiology and consequences of obesity . Obesity Res. 2002 ; 10 : 97S – 104S . Google Scholar Crossref Search ADS WorldCat 6 National Institutes of Health, Office of Medical Applications of Research . Health Implications of Obesity. Vol 5. US Department of Health and Human Services, Public Health Service, National Institutes of Health, Office of Medical Applications of Research; 1985 . 7 Poirier P Giles TD Bray GA , et al. . Obesity and cardiovascular disease: pathophysiology, evaluation, and effect of weight loss . Circulation. 2006 ; 113 : 898 – 918 . Google Scholar Crossref Search ADS PubMed WorldCat 8 Furukawa S Fujita T Shimabukuro M , et al. . Increased oxidative stress in obesity and its impact on metabolic syndrome . J Clin Invest. 2004 ; 114 : 1752 – 1761 . Google Scholar Crossref Search ADS PubMed WorldCat 9 Thomas EL Frost G Taylor-Robinson SD , et al. . Excess body fat in obese and normal-weight subjects . Nutr Res Rev. 2012 ; 25 : 150 – 161 . Google Scholar Crossref Search ADS PubMed WorldCat 10 Sanyal D Mukhopadhyay P Pandit K , et al. . Central obesity but not generalised obesity (body mass index) predicts high prevalence of fatty liver (NRFLD), in recently detected untreated, IGT and type 2 diabetes Indian subjects . J Indian Med Assoc. 2009 ; 107 : 755 – 758 . Google Scholar PubMed OpenURL Placeholder Text WorldCat 11 Carmienke S Freitag MH Pischon T , et al. . General and abdominal obesity parameters and their combination in relation to mortality: a systematic review and meta-regression analysis . Eur J Clin Nutr. 2013 ; 67 : 573 – 585 . Google Scholar Crossref Search ADS PubMed WorldCat 12 Song X Jousilahti P Stehouwer CD , et al. . Comparison of various surrogate obesity indicators as predictors of cardiovascular mortality in four European populations . Eur J Clin Nutr. 2013 ; 67 : 1298 – 1302 . Google Scholar Crossref Search ADS PubMed WorldCat 13 Bila WC Freitas AE Galdino AS , et al. . Deuterium oxide dilution and body composition in overweight and obese schoolchildren aged 6-9 years . Jornal de Pediatria. 2016 ; 92 : 46 – 52 . Google Scholar Crossref Search ADS PubMed WorldCat 14 Smalley KJ Knerr AN Kendrick ZV , et al. . Reassessment of body mass indices . Am J Clin Nutr. 1990 ; 52 : 405 – 408 . Google Scholar Crossref Search ADS PubMed WorldCat 15 Goh VH Tain C Tong TY , et al. . Are BMI and other anthropometric measures appropriate as indices for obesity? A study in an Asian population . J Lipid Res. 2004 ; 45 : 1892 – 1898 . Google Scholar Crossref Search ADS PubMed WorldCat 16 Romero-Corral A Somers VK Sierra-Johnson J , et al. . Accuracy of body mass index in diagnosing obesity in the adult general population . Int J Obes. 2008 ; 32 : 959 – 966 . Google Scholar Crossref Search ADS WorldCat 17 Bellizzi MC Dietz WH. Workshop on childhood obesity: summary of the discussion . Am J Clin Nutr . 1999 ; 70 : S173 – S175 . Google Scholar Crossref Search ADS WorldCat 18 Burkhauser RV Cawley J. Beyond BMI: the value of more accurate measures of fatness and obesity in social science research . J Health Econ. 2008 ; 27 : 519 – 529 . Google Scholar Crossref Search ADS PubMed WorldCat 19 Rothman KJ. BMI-related errors in the measurement of obesity . Int J Obes. 2008 ; 32 : S56 – S59 . Google Scholar Crossref Search ADS WorldCat 20 Gelber RP Gaziano JM Orav EJ , et al. . Measures of obesity and cardiovascular risk among men and women . J Am Coll Cardiol. 2008 ; 52 : 605 – 615 . Google Scholar Crossref Search ADS PubMed WorldCat 21 Molarius A Seidell J. Selection of anthropometric indicators for classification of abdominal fatness – a critical review . Int J Obes. 1998 ; 22 : 719 – 727 . Google Scholar Crossref Search ADS WorldCat 22 Burniat W Cole T Lissau I , et al. , eds. Child and Adolescent Obesity: Causes and Consequences, Prevention and Management. Vol. 1 . Cambridge: Cambridge University Press ; 2002 . Google Scholar Crossref Search ADS Google Scholar Google Preview WorldCat COPAC 23 Moher D Liberati A Tetzlaff J , et al. . Preferred reporting items for systematic reviews and meta-analyses: the PRISMA statement . BMJ. 2009 ; 339 : b2535 . Google Scholar Crossref Search ADS PubMed WorldCat 24 Abulmeaty MM Almajwal AM Almadani NK , et al. . Anthropometric and central obesity indices as predictors of long-term cardiometabolic risk among Saudi young and middle-aged men and women . Saudi Med J. 2017 ; 38 : 372 – 380 . Google Scholar Crossref Search ADS PubMed WorldCat 25 Al-Gindan YY Hankey CR Govan L , et al. . Derivation and validation of simple anthropometric equations to predict adipose tissue mass and total fat mass with MRI as the reference method . Br J Nutr. 2015 ; 114 : 1852 – 1867 . Google Scholar Crossref Search ADS PubMed WorldCat 26 Carlsson AC Riserus U Engstrom G , et al. . Novel and established anthropometric measures and the prediction of incident cardiovascular disease: a cohort study . Int J Obes. 2013 ; 37 : 1579 – 1585 . Google Scholar Crossref Search ADS WorldCat 27 Chagas P Caramori P Barcellos C , et al. . Association of different anthropometric measures and indices with coronary atherosclerotic burden . Arq Bras Cardiol. 2011 ; 97 : 397 – 401 . Google Scholar Crossref Search ADS PubMed WorldCat 28 Chang Y Guo X Chen Y , et al. . A body shape index and body roundness index: two new body indices to identify diabetes mellitus among rural populations in northeast China . BMC Public Health. 2015 ; 15 : 794. Google Scholar Crossref Search ADS PubMed WorldCat 29 Chung W Park CG Ryu OH. Association of a new measure of obesity with hypertension and health-related quality of life . PloS One. 2016 ; 11 : e0155399. Google Scholar Crossref Search ADS PubMed WorldCat 30 Fu S Luo L Ye P , et al. . The abilities of new anthropometric indices in identifying cardiometabolic abnormalities, and influence of residence area and lifestyle on these anthropometric indices in a Chinese community-dwelling population . Clin Interv Aging. 2014 ; 9 : 179 – 189 . Google Scholar Crossref Search ADS PubMed WorldCat 31 Haghighatdoost F Sarrafzadegan N Mohammadifard N , et al. . Assessing body shape index as a risk predictor for cardiovascular diseases and metabolic syndrome among Iranian adults . Nutrition . 2014 ; 30 : 636 – 644 . Google Scholar Crossref Search ADS PubMed WorldCat 32 Jung KJ Kimm H Yun JE , et al. . Thigh circumference and diabetes: obesity as a potential effect modifier . J Epidemiol. 2013 ; 23 : 329 – 336 . Google Scholar Crossref Search ADS PubMed WorldCat 33 Kaneko R Nakazaki N Tagawa T , et al. . A new index of abdominal obesity which effectively predicts risk of colon tumor development in female Japanese . Asian Pac J Cancer Prevent . 2014 ; 15 : 1005 – 1010 . Google Scholar Crossref Search ADS WorldCat 34 Katulanda P Jayawardena MA Sheriff MH , et al. . The distance between the lower edge of the xiphisternum and the center of the umbilicus as an indicator of abdominal obesity and cardiovascular disease risk . Obes Facts. 2010 ; 3 : 201 – 204 . Google Scholar Crossref Search ADS PubMed WorldCat 35 Kommuri NV Zalawadiya SK Veeranna V , et al. . Association between various anthropometric measures of obesity and markers of subclinical atherosclerosis . Expert Rev Cardiovasc Ther . 2016 ; 14 : 127 – 135 . Google Scholar Crossref Search ADS PubMed WorldCat 36 Krakauer NY Krakauer JC. An anthropometric risk index based on combining height, weight, waist, and hip measurements . J Obes . 2016 ; 2016 : 1. Google Scholar Crossref Search ADS WorldCat 37 Li C Ford ES Zhao G , et al. . Waist-to-thigh ratio and diabetes among US adults: the Third National Health and Nutrition Examination Survey . Diabetes Res Clin Pract. 2010 ; 89 : 79 – 87 . Google Scholar Crossref Search ADS PubMed WorldCat 38 Motamed N Perumal D Zamani F , et al. . Conicity index and waist-to-hip ratio are superior obesity indices in predicting 10-year cardiovascular risk among men and women . Clin Cardiol. 2015 ; 38 : 527 – 534 . Google Scholar Crossref Search ADS PubMed WorldCat 39 Ononamadu CJ Ezekwesili CN Onyeukwu OF , et al. . Comparative analysis of anthropometric indices of obesity as correlates and potential predictors of risk for hypertension and prehypertension in a population in Nigeria . Cardiovasc J Afr. 2017 ; 28 : 92 – 99 . Google Scholar Crossref Search ADS PubMed WorldCat 40 Rahman SA Adjeroh D. Surface-based body shape index and its relationship with all-cause mortality . PLoS One. 2015 ; 10 : e0144639. Google Scholar Crossref Search ADS PubMed WorldCat 41 Vogel P Stein A Marcadenti A. Visceral adiposity index and prognosis among patients with ischemic heart failure . Sao Paulo Med J. 2016 ; 134 : 211 – 218 . Google Scholar Crossref Search ADS PubMed WorldCat 42 Zhou JY Ge H Zhu MF , et al. . Neck circumference as an independent predictive contributor to cardio-metabolic syndrome . Cardiovasc Diabetol. 2013 ; 12 : 76. Google Scholar Crossref Search ADS PubMed WorldCat 43 Assyov Y Gateva A Tsakova A , et al. . A comparison of the clinical usefulness of neck circumference and waist circumference in individuals with severe obesity . Endocrine Res . 2017 ; 42 : 6 – 14 . Google Scholar Crossref Search ADS WorldCat 44 Akin Y Gulmez H Savas M , et al. . Relationship between neck circumference and overactive bladder in women with metabolic syndrome: a preliminary study . Wien Klin Wochenschr. 2016 ; 128 : 581 – 586 . Google Scholar Crossref Search ADS PubMed WorldCat 45 Coelho HJJ Sampaio RA Goncalvez IO , et al. . Cutoffs and cardiovascular risk factors associated with neck circumference among community-dwelling elderly adults: a cross-sectional study . Sao Paulo Med J. 2016 ; 134 : 519 – 527 . Google Scholar Crossref Search ADS PubMed WorldCat 46 Joshipura K Munoz-Torres F Vergara J , et al. . Neck circumference may be a better alternative to standard anthropometric measures . J Diabetes Res. 2016 ; 2016 : 1. Google Scholar Crossref Search ADS WorldCat 47 Ozkaya I Tunckale A. Neck circumference positively related with central obesity and overweight in Turkish university students: a preliminary study . Centr Eur J Public Health 2016 ; 24 : 91 – 94 . Google Scholar Crossref Search ADS WorldCat 48 Yang GR Yuan SY Fu HJ , et al. . Neck circumference positively related with central obesity, overweight, and metabolic syndrome in Chinese subjects with type 2 diabetes: Beijing Community Diabetes Study 4 . Diabetes Care. 2010 ; 33 : 2465 – 2467 . Google Scholar Crossref Search ADS PubMed WorldCat 49 Anunciacao PC Ribeiro RC Pereira MQ , et al. . Different measurements of waist circumference and sagittal abdominal diameter and their relationship with cardiometabolic risk factors in elderly men . J Hum Nutr Diet. 2014 ; 27 : 162 – 167 . Google Scholar Crossref Search ADS PubMed WorldCat 50 Dahlen EM Bjarnegard N Lanne T , et al. . Sagittal abdominal diameter is a more independent measure compared with waist circumference to predict arterial stiffness in subjects with type 2 diabetes – a prospective observational cohort study . Cardiovasc Diabetol. 2013 ; 12 : 55. Google Scholar Crossref Search ADS PubMed WorldCat 51 Dahlen EM Tengblad A Lanne T , et al. . Abdominal obesity and low-grade systemic inflammation as markers of subclinical organ damage in type 2 diabetes . Diabetes Metab. 2014 ; 40 : 76 – 81 . Google Scholar Crossref Search ADS PubMed WorldCat 52 Duarte Pimentel G Portero-McLellan KC Maestá N , et al. . Accuracy of sagittal abdominal diameter as predictor of abdominal fat among Brazilian adults: a comparation with waist circumference . Nutr Hosp. 2010 ; 25 : 656 – 661 . Google Scholar PubMed OpenURL Placeholder Text WorldCat 53 Pekkarinen E Vanninen E Lansimies E , et al. . Relation between body composition, abdominal obesity, and lung function . Clin Physiol Funct Imaging. 2012 ; 32 : 83 – 88 . Google Scholar Crossref Search ADS PubMed WorldCat 54 Radholm K Tengblad A Dahlen E , et al. . The impact of using sagittal abdominal diameter to predict major cardiovascular events in European patients with type 2 diabetes . Nutr Metab Cardiovasc Dis. 2017 ; 27 : 418 – 422 . Google Scholar Crossref Search ADS PubMed WorldCat 55 Carlsson AC Riserus U Arnlov J , et al. . Prediction of cardiovascular disease by abdominal obesity measures is dependent on body weight and sex – results from two community based cohort studies . Nutr Metab Cardiovasc Dis . 2014 ; 24 : 891 – 899 . Google Scholar Crossref Search ADS PubMed WorldCat 56 Theorell-Haglow J Berne C Janson C , et al. . Associations between short sleep duration and central obesity in women . Sleep. 2010 ; 33 : 593 – 598 . Google Scholar Crossref Search ADS PubMed WorldCat 57 Vasques AC Souza JR Yamanaka A , et al. . Sagittal abdominal diameter as a marker for epicardial adipose tissue in premenopausal women . Metabolism . 2013 ; 62 : 1032 – 1036 . Google Scholar Crossref Search ADS PubMed WorldCat 58 Heitmann BL Frederiksen P. Thigh circumference and risk of heart disease and premature death: prospective cohort study . BMJ. 2009 ; 339 :b3292. OpenURL Placeholder Text WorldCat 59 Snijder MB Dekker JM Visser M , et al. . Larger thigh and hip circumferences are associated with better glucose tolerance: the Hoorn study . Obesity Res. 2003 ; 11 : 104 – 111 . Google Scholar Crossref Search ADS WorldCat 60 Fay RA Dey PL Saadie CM , et al. . Ponderal index: a better definition of the ‘at risk’ group with intrauterine growth problems than birth-weight for gestational age in term infants . Aust N Z J Obstet Gynaecol. 1991 ; 31 : 17 – 19 . Google Scholar Crossref Search ADS PubMed WorldCat 61 Rosenblad A Leppert J Nilsson G. The waist-hip-height-ratio (WHHR): an improvement of the waist-hip-ratio for predicting all-cause mortality . J Diabetes. 2011 ; 3 : 1753 . OpenURL Placeholder Text WorldCat 62 Hardy DS Stallings DT Garvin JT , et al. . Best anthropometric discriminators of incident type 2 diabetes among white and black adults: a longitudinal ARIC study . PLoS One. 2017 ; 12 :e0168282 . OpenURL Placeholder Text WorldCat 63 Valdez R. A simple model-based index of abdominal adiposity . J Clin Epidemiol. 1991 ; 44 : 955 – 956 . Google Scholar Crossref Search ADS PubMed WorldCat 64 Almeida RT Almeida MM Araujo TM. Abdominal obesity and cardiovascular risk: performance of anthropometric indexes in women . Arq Bras Cardiol. 2009 ; 92 : 345 – 350 , 362–367, 375–380. Google Scholar Crossref Search ADS PubMed WorldCat 65 Chakraborty R Bose K. Comparison of body adiposity indices in predicting blood pressure and hypertension among slum-dwelling men in Kolkata . Malays J Nutr . 2012 ; 18 : 319 – 328 . Google Scholar PubMed OpenURL Placeholder Text WorldCat 66 Cheah WL Hazmi H Chia HQ , et al. . Hypertension and its association with anthropometric indexes among pre-university students . Int J Adolesc Med Health . 2016 ; 28 : 373 – 379 . Google Scholar Crossref Search ADS PubMed WorldCat 67 Costa EC Sa JC Soares EM , et al. . Anthropometric indices of central obesity how discriminators of metabolic syndrome in Brazilian women with polycystic ovary syndrome . Gynecol Endocrinol. 2012 ; 28 : 12 – 15 . Google Scholar Crossref Search ADS PubMed WorldCat 68 Wang H Liu A Zhao T , et al. . Comparison of anthropometric indices for predicting the risk of metabolic syndrome and its components in Chinese adults: a prospective, longitudinal study . BMJ Open. 2017 ; 7 : e016062. Google Scholar Crossref Search ADS PubMed WorldCat 69 Ruperto M Barril G Sanchez-Muniz FJ. Usefulness of the conicity index together with the conjoint use of adipocytokines and nutritional-inflammatory markers in hemodialysis patients . J Physiol Biochem. 2017 ; 73 : 67 – 75 . Google Scholar Crossref Search ADS PubMed WorldCat 70 Bergman RN Stefanovski D Buchanan TA , et al. . A better index of body adiposity . Obesity (Silver Spring). 2011 ; 19 : 1083 – 1089 . Google Scholar Crossref Search ADS PubMed WorldCat 71 Belarmino G Horie LM Sala PC , et al. . Body adiposity index performance in estimating body fat in a sample of severely obese Brazilian patients . Nutr J. 2015 ; 14 : 130. Google Scholar Crossref Search ADS PubMed WorldCat 72 Ramirez-Velez R Correa-Bautista JE Gonzalez-Ruiz K , et al. . Predictive validity of the body adiposity index in overweight and obese adults using dual-energy x-ray absorptiometry . Nutrients. 2016 ; 8 : 737 . Google Scholar Crossref Search ADS WorldCat 73 Santos DA Silva AM Matias CN , et al. . Utility of novel body indices in predicting fat mass in elite athletes . Nutrition . 2015 ; 31 : 948 – 954 . Google Scholar Crossref Search ADS PubMed WorldCat 74 D’Elia L Manfredi M Sabino P , et al. . The Olivetti Heart Study: predictive value of a new adiposity index on risk of hypertension, blood pressure, and subclinical organ damage . Nutr Metab Cardiovasc Dis . 2016 ; 26 : 630 – 636 . Google Scholar Crossref Search ADS PubMed WorldCat 75 Lam BC Koh GC Chen C , et al. . Comparison of Body Mass Index (BMI), Body Adiposity Index (BAI), Waist Circumference (WC), Waist-To-Hip Ratio (WHR) and Waist-To-Height Ratio (WHtR) as predictors of cardiovascular disease risk factors in an adult population in Singapore . PLoS One. 2015 ; 10 : e0122985. Google Scholar Crossref Search ADS PubMed WorldCat 76 Liu PJ Ma F Lou HP , et al. . Body roundness index and body adiposity index: two new anthropometric indices to identify metabolic syndrome among Chinese postmenopausal women . Climacteric . 2016 ; 19 : 433 – 439 . Google Scholar Crossref Search ADS PubMed WorldCat 77 Garcia AI Nino-Silva LA Gonzalez-Ruiz K , et al. . Body adiposity index as marker of obesity and cardiovascular risk in adults from Bogotá, Colombia . Endocrinol Nutr. 2015 ; 62 : 130 – 137 . Google Scholar Crossref Search ADS PubMed WorldCat 78 Talaei M Sadeghi M Marshall T , et al. . Anthropometric indices predicting incident type 2 diabetes in an Iranian population: the Isfahan Cohort Study . Diabetes Metab. 2013 ; 39 : 424 – 431 . Google Scholar Crossref Search ADS PubMed WorldCat 79 Thomas DM Bredlau C Bosy-Westphal A , et al. . Relationships between body roundness with body fat and visceral adipose tissue emerging from a new geometrical model . Obesity (Silver Spring) . 2013 ; 21 : 2264 – 2271 . Google Scholar Crossref Search ADS PubMed WorldCat 80 Martin V Davila-Batista V Castilla J , et al. . Comparison of body mass index (BMI) with the CUN-BAE body adiposity estimator in the prediction of hypertension and type 2 diabetes . BMC Public Health . 2016 ; 16 : 82. Google Scholar Crossref Search ADS PubMed WorldCat 81 Gomez-Ambrosi J Silva C Catalan V , et al. . Clinical usefulness of a new equation for estimating body fat . Diabetes Care. 2012 ; 35 : 383 – 388 . Google Scholar Crossref Search ADS PubMed WorldCat 82 Vinknes KJ Nurk E Tell GS , et al. . The relation of CUN-BAE index and BMI with body fat, cardiovascular events and diabetes during a 6-year follow-up: the Hordaland Health Study . Clin Epidemiol. 2017 ; 9 : 555 – 566 . Google Scholar Crossref Search ADS PubMed WorldCat 83 Gurney JM Jelliffe DB. Arm anthropometry in nutritional assessment: nomogram for rapid calculation of muscle circumference and cross-sectional muscle and fat areas . Am J Clin Nutr . 1973 ; 26 : 912 – 915 . Google Scholar Crossref Search ADS PubMed WorldCat 84 Biolo G Di Girolamo FG Breglia A , et al. . Inverse relationship between “a body shape index” (ABSI) and fat-free mass in women and men: insights into mechanisms of sarcopenic obesity . Clin Nutr . 2015 ; 34 : 323 – 327 . Google Scholar Crossref Search ADS PubMed WorldCat 85 Bertoli S Leone A Krakauer NY , et al. . Association of Body Shape Index (ABSI) with cardio-metabolic risk factors: a cross-sectional study of 6081 Caucasian adults . PLoS One. 2017 ; 12 : e0185013. Google Scholar Crossref Search ADS PubMed WorldCat 86 Grant JF Chittleborough CR Shi Z , et al. . The association between A Body Shape Index and mortality: results from an Australian cohort . PLoS One. 2017 ; 12 : e0181244. Google Scholar Crossref Search ADS PubMed WorldCat 87 Krakauer NY Krakauer JC. A new body shape index predicts mortality hazard independently of body mass index . PLoS One. 2012 ; 7 : e39504. Google Scholar Crossref Search ADS PubMed WorldCat 88 Krakauer NY Krakauer JC. Dynamic association of mortality hazard with body shape . PLoS One. 2014 ; 9 : e88793. Google Scholar Crossref Search ADS PubMed WorldCat 89 Guerrero-Romero F Rodríguez-Morán M. Abdominal volume index. An anthropometry-based index for estimation of obesity is strongly related to impaired glucose tolerance and type 2 diabetes mellitus . Arch Med Res . 2003 ; 34 : 428 – 432 . Google Scholar Crossref Search ADS PubMed WorldCat 90 Neufeld LM Jones-Smith JC Garcia R , et al. . Anthropometric predictors for the risk of chronic disease in non-diabetic, non-hypertensive young Mexican women . Public Health Nutr. 2008 ; 11 : 159 – 167 . Google Scholar Crossref Search ADS PubMed WorldCat 91 Duarte-Rojo A Sosa-Lozano LA Saul A , et al. . Methods for measuring abdominal obesity in the prediction of severe acute pancreatitis, and their correlation with abdominal fat areas assessed by computed tomography . Aliment Pharmacol Ther . 2010 ; 32 : 244 – 253 . Google Scholar Crossref Search ADS PubMed WorldCat 92 Lu B Zhou J Waring ME , et al. . Abdominal obesity and peripheral vascular disease in men and women: a comparison of waist-to-thigh ratio and waist circumference as measures of abdominal obesity . Atherosclerosis . 2010 ; 208 : 253 – 257 . Google Scholar Crossref Search ADS PubMed WorldCat 93 Kumar S Kumar K Bajaj S , et al. . Waist-thigh ratio: a surrogate marker for type 2 diabetes mellitus in Asian North Indian patients . Indian J Endocrinol Metab. 2018 ; 22 : 47 – 49 . Google Scholar Crossref Search ADS PubMed WorldCat 94 Kahn HS Austin H Williamson DF , et al. . Simple anthropometric indices associated with ischemic heart disease . J Clin Epidemiol . 1996 ; 49 : 1017 – 1024 . Google Scholar Crossref Search ADS PubMed WorldCat 95 Gately PJ Radley D Cooke CB , et al. . Comparison of body composition methods in overweight and obese children . J Appl Physiol . 2003 ; 95 : 2039 – 2046 . Google Scholar Crossref Search ADS PubMed WorldCat 96 Misra A Khurana L. Obesity-related non-communicable diseases: South Asians vs White Caucasians . Int J Obes. 2011 ; 35 : 167 – 187 . Google Scholar Crossref Search ADS WorldCat 97 Katulanda P Jayawardena MA Sheriff MH , et al. . Derivation of anthropometric cut-off levels to define CVD risk in Sri Lankan adults . Br J Nutr. 2011 ; 105 : 1084 – 1090 . Google Scholar Crossref Search ADS PubMed WorldCat 98 Gray LJ Yates T Davies MJ , et al. . Defining obesity cut-off points for migrant South Asians . PloS One. 2011 ; 6 : e26464. Google Scholar Crossref Search ADS PubMed WorldCat 99 Gasteyger C Tremblay A. Metabolic impact of body fat distribution . J Endocrinol Invest. 2002 ; 25 : 876 – 883 . Google Scholar Crossref Search ADS PubMed WorldCat 100 Britton KA Massaro JM Murabito JM , et al. . Body fat distribution, incident cardiovascular disease, cancer, and all-cause mortality . J Am Coll Cardiol . 2013 ; 62 : 921 – 925 . Google Scholar Crossref Search ADS PubMed WorldCat 101 Liu J Fox CS Hickson DA , et al. . Impact of abdominal visceral and subcutaneous adipose tissue on cardiometabolic risk factors: the Jackson Heart Study . J Clin Endocrinol Metab. 2010 ; 95 : 5419 – 5426 . Google Scholar Crossref Search ADS PubMed WorldCat 102 Abramowitz MK Hall CB Amodu A , et al. . Muscle mass, BMI, and mortality among adults in the United States: a population-based cohort study . PLoS One. 2018 ; 13 : e0194697. Google Scholar Crossref Search ADS PubMed WorldCat 103 Prado CM Gonzalez MC Heymsfield SB. Body composition phenotypes and obesity paradox . Curr Opin Clin Nutr Metab Care. 2015 ; 18 : 535 – 551 . Google Scholar Crossref Search ADS PubMed WorldCat © The Author(s) 2019. Published by Oxford University Press on behalf of the International Life Sciences Institute. All rights reserved. For permissions, please e-mail: [email protected]. This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/open_access/funder_policies/chorus/standard_publication_model)